Publications

2023

  • D. M. Karine Miras and A. E. Eiben, “Hu-bot: promoting the cooperation between humans and mobile robots.” 2023.
    [BibTeX] [Abstract] [Download PDF]

    This paper investigates human–robot collaboration in a novel setup: a human helps a mobile robot that can move and navigate freely in an environment. Specifically, the human helps by remotely taking over control during the learning of a task. The task is to find and collect several items in a walled arena, and Reinforcement Learning is used to seek a suitable controller. If the human observes undesired robot behavior, they can directly issue commands for the wheels through a game joystick. Experiments in a simulator showed that human assistance improved robot behavior efficacy by 30% and efficiency by 12%. The best policies were also tested in real life, using physical robots. Hardware experiments showed no significant difference concerning the simulations, providing empirical validation of our approach in practice.

    @inproceedings{miras23hubot,
    title={Hu-bot: promoting the cooperation between humans and mobile robots},
    author={Karine Miras, Decebal Mocanu & A. E. Eiben },
    journal={Neural Computing and Applications },
    year={2023},
    url = {https://link.springer.com/article/10.1007/s00521-022-08061-z},
    abstract = { This paper investigates human–robot collaboration in a novel setup: a human helps a mobile robot that can move and navigate freely in an environment. Specifically, the human helps by remotely taking over control during the learning of a task. The task is to find and collect several items in a walled arena, and Reinforcement Learning is used to seek a suitable controller. If the human observes undesired robot behavior, they can directly issue commands for the wheels through a game joystick. Experiments in a simulator showed that human assistance improved robot behavior efficacy by 30% and efficiency by 12%. The best policies were also tested in real life, using physical robots. Hardware experiments showed no significant difference concerning the simulations, providing empirical validation of our approach in practice.}
    }

  • T. Kim, M. Cochez, V. François-Lavet, M. Neerincx, and P. Vossen, “A Machine with Short-Term, Episodic, and Semantic Memory Systems.” 2023. doi:10.48550/ARXIV.2212.02098
    [BibTeX] [Abstract] [Download PDF]

    Inspired by the cognitive science theory of the explicit human memory systems, we have modeled an agent with short-term, episodic, and semantic memory systems, each of which is modeled with a knowledge graph. To evaluate this system and analyze the behavior of this agent, we designed and released our own reinforcement learning agent environment, “the Room”, where an agent has to learn how to encode, store, and retrieve memories to maximize its return by answering questions. We show that our deep Q-learning based agent successfully learns whether a short-term memory should be forgotten, or rather be stored in the episodic or semantic memory systems. Our experiments indicate that an agent with human-like memory systems can outperform an agent without this memory structure in the environment.

    @InProceedings{https://doi.org/10.48550/arxiv.2212.02098,
    title = {A Machine with Short-Term, Episodic, and Semantic
    Memory Systems},
    author = {Kim, Taewoon and Cochez, Michael and François-Lavet,
    Vincent and Neerincx, Mark and Vossen, Piek},
    doi = {10.48550/ARXIV.2212.02098},
    url = {https://arxiv.org/abs/2212.02098},
    year = {2023},
    month = {Feb.},
    journal = {Proceedings of the AAAI Conference on Artificial
    Intelligence},
    abstract = {Inspired by the cognitive science theory of the
    explicit human memory systems, we have modeled an
    agent with short-term, episodic, and semantic memory
    systems, each of which is modeled with a knowledge
    graph. To evaluate this system and analyze the
    behavior of this agent, we designed and released our
    own reinforcement learning agent environment, "the
    Room", where an agent has to learn how to encode,
    store, and retrieve memories to maximize its return
    by answering questions. We show that our deep
    Q-learning based agent successfully learns whether a
    short-term memory should be forgotten, or rather be
    stored in the episodic or semantic memory
    systems. Our experiments indicate that an agent with
    human-like memory systems can outperform an agent
    without this memory structure in the environment.},
    }

2022

  • M. Tsfasman, K. Fenech, M. Tarvirdians, A. Lorincz, C. Jonker, and C. Oertel, “Towards Creating a Conversational Memory for Long-Term Meeting Support: Predicting Memorable Moments in Multi-Party Conversations through Eye-Gaze,” in Proceedings of the 2022 International Conference on Multimodal Interaction, New York, NY, USA, 2022, p. 94–104. doi:10.1145/3536221.3556613
    [BibTeX] [Abstract] [Download PDF]

    When working in a group, it is essential to understand each other’s viewpoints to increase group cohesion and meeting productivity. This can be challenging in teams: participants might be left misunderstood and the discussion could be going around in circles. To tackle this problem, previous research on group interactions has addressed topics such as dominance detection, group engagement, and group creativity. Conversational memory, however, remains a widely unexplored area in the field of multimodal analysis of group interaction. The ability to track what each participant or a group as a whole find memorable from each meeting would allow a system or agent to continuously optimise its strategy to help a team meet its goals. In the present paper, we therefore investigate what participants take away from each meeting and how it is reflected in group dynamics.As a first step toward such a system, we recorded a multimodal longitudinal meeting corpus (MEMO), which comprises a first-party annotation of what participants remember from a discussion and why they remember it. We investigated whether participants of group interactions encode what they remember non-verbally and whether we can use such non-verbal multimodal features to predict what groups are likely to remember automatically. We devise a coding scheme to cluster participants’ memorisation reasons into higher-level constructs. We find that low-level multimodal cues, such as gaze and speaker activity, can predict conversational memorability. We also find that non-verbal signals can indicate when a memorable moment starts and ends. We could predict four levels of conversational memorability with an average accuracy of 44 \%. We also showed that reasons related to participants’ personal feelings and experiences are the most frequently mentioned grounds for remembering meeting segments.

    @inproceedings{10.1145/3536221.3556613,
    author = {Tsfasman, Maria and Fenech, Kristian and Tarvirdians, Morita and Lorincz, Andras and Jonker, Catholijn and Oertel, Catharine},
    title = {Towards Creating a Conversational Memory for Long-Term Meeting Support: Predicting Memorable Moments in Multi-Party Conversations through Eye-Gaze},
    year = {2022},
    isbn = {9781450393904},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    url = {https://doi.org/10.1145/3536221.3556613},
    doi = {10.1145/3536221.3556613},
    abstract = {When working in a group, it is essential to understand each other’s viewpoints to increase group cohesion and meeting productivity. This can be challenging in teams: participants might be left misunderstood and the discussion could be going around in circles. To tackle this problem, previous research on group interactions has addressed topics such as dominance detection, group engagement, and group creativity. Conversational memory, however, remains a widely unexplored area in the field of multimodal analysis of group interaction. The ability to track what each participant or a group as a whole find memorable from each meeting would allow a system or agent to continuously optimise its strategy to help a team meet its goals. In the present paper, we therefore investigate what participants take away from each meeting and how it is reflected in group dynamics.As a first step toward such a system, we recorded a multimodal longitudinal meeting corpus (MEMO), which comprises a first-party annotation of what participants remember from a discussion and why they remember it. We investigated whether participants of group interactions encode what they remember non-verbally and whether we can use such non-verbal multimodal features to predict what groups are likely to remember automatically. We devise a coding scheme to cluster participants’ memorisation reasons into higher-level constructs. We find that low-level multimodal cues, such as gaze and speaker activity, can predict conversational memorability. We also find that non-verbal signals can indicate when a memorable moment starts and ends. We could predict four levels of conversational memorability with an average accuracy of 44 \%. We also showed that reasons related to participants’ personal feelings and experiences are the most frequently mentioned grounds for remembering meeting segments.},
    booktitle = {Proceedings of the 2022 International Conference on Multimodal Interaction},
    pages = {94–104},
    numpages = {11},
    keywords = {multi-party interaction, social signals, conversational memory, multi-modal corpora},
    location = {Bengaluru, India},
    series = {ICMI '22}
    }

  • B. Dudzik, D. Küster, D. St-Onge, and F. Putze, “The 4th Workshop on Modeling Socio-Emotional and Cognitive Processes from Multimodal Data In-the-Wild (MSECP-Wild),” in Proceedings of the 2022 International Conference on Multimodal Interaction, New York, NY, USA, 2022, p. 803–804. doi:10.1145/3536221.3564029
    [BibTeX] [Abstract] [Download PDF]

    The ability to automatically infer relevant aspects of human users’ thoughts and feelings is crucial for technologies to adapt their behaviors in complex interactions intelligently (e.g., social robots or tutoring systems). Research on multimodal analysis has demonstrated the potential of technology to provide such estimates for a broad range of internal states and processes. However, constructing robust enough approaches for deployment in real-world applications remains an open problem. The MSECP-Wild workshop series serves as a multidisciplinary forum to present and discuss research addressing this challenge. This 4th iteration focuses on addressing varying contextual conditions (e.g., throughout an interaction or across different situations and environments) in intelligent systems as a crucial barrier for more valid real-world predictions and actions. Submissions to the workshop span efforts relevant to multimodal data collection and context-sensitive modeling. These works provide important impulses for discussions of the state-of-the-art and opportunities for future research on these subjects.

    @inproceedings{10.1145/3536221.3564029,
    author = {Dudzik, Bernd and K\"{u}ster, Dennis and St-Onge, David and Putze, Felix},title = {The 4th Workshop on Modeling Socio-Emotional and Cognitive Processes from Multimodal Data In-the-Wild (MSECP-Wild)},
    year = {2022},
    isbn = {9781450393904},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    url = {https://pure.tudelft.nl/admin/files/140622118/3536221.3564029.pdf},
    doi = {10.1145/3536221.3564029},
    abstract = {The ability to automatically infer relevant aspects of human users’ thoughts and feelings is crucial for technologies to adapt their behaviors in complex interactions intelligently (e.g., social robots or tutoring systems). Research on multimodal analysis has demonstrated the potential of technology to provide such estimates for a broad range of internal states and processes. However, constructing robust enough approaches for deployment in real-world applications remains an open problem. The MSECP-Wild workshop series serves as a multidisciplinary forum to present and discuss research addressing this challenge. This 4th iteration focuses on addressing varying contextual conditions (e.g., throughout an interaction or across different situations and environments) in intelligent systems as a crucial barrier for more valid real-world predictions and actions. Submissions to the workshop span efforts relevant to multimodal data collection and context-sensitive modeling. These works provide important impulses for discussions of the state-of-the-art and opportunities for future research on these subjects.},
    booktitle = {Proceedings of the 2022 International Conference on Multimodal Interaction},
    pages = {803–804},
    numpages = {2},
    keywords = {Context-awareness, User-Modeling, Multimodal Data, Social Signal Processing, Affective Computing, Ubiquitous Computing},
    location = {Bengaluru, India},
    series = {ICMI '22}
    }

  • B. Dudzik and H. Hung, “Exploring the Detection of Spontaneous Recollections during Video-Viewing In-the-Wild Using Facial Behavior Analysis,” in Proceedings of the 2022 International Conference on Multimodal Interaction, New York, NY, USA, 2022, p. 236–246. doi:10.1145/3536221.3556609
    [BibTeX] [Abstract] [Download PDF]

    Intelligent systems might benefit from automatically detecting when a stimulus has triggered a user’s recollection of personal memories, e.g., to identify that a piece of media content holds personal significance for them. While computational research has demonstrated the potential to identify related states based on facial behavior (e.g., mind-wandering), the automatic detection of spontaneous recollections specifically has not been investigated this far. Motivated by this, we present machine learning experiments exploring the feasibility of detecting whether a video clip has triggered personal memories in a viewer based on the analysis of their Head Rotation, Head Position, Eye Gaze, and Facial Expressions. Concretely, we introduce an approach for automatic detection and evaluate its potential for predictions using in-the-wild webcam recordings. Overall, our findings demonstrate the capacity for above chance detections in both settings, with substantially better performance for the video-independent variant. Beyond this, we investigate the role of person-specific recollection biases for predictions of our video-independent models and the importance of specific modalities of facial behavior. Finally, we discuss the implications of our findings for detecting recollections and user-modeling in adaptive systems.

    @inproceedings{10.1145/3536221.3556609,
    author = {Dudzik, Bernd and Hung, Hayley},
    title = {Exploring the Detection of Spontaneous Recollections during Video-Viewing In-the-Wild Using Facial Behavior Analysis},
    year = {2022},
    isbn = {9781450393904},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    url = {https://dl.acm.org/doi/10.1145/3536221.3556609},
    doi = {10.1145/3536221.3556609},
    abstract = {Intelligent systems might benefit from automatically detecting when a stimulus has triggered a user’s recollection of
    personal memories, e.g., to identify that a piece of media content holds personal significance for them. While computational research has demonstrated the potential to identify related states based on facial behavior (e.g., mind-wandering), the automatic detection of spontaneous recollections specifically has not been investigated this far. Motivated by this, we present machine learning experiments exploring the feasibility of detecting whether a video clip has triggered personal memories in a viewer based on the analysis of their Head Rotation, Head Position, Eye Gaze, and Facial Expressions. Concretely, we introduce an approach for automatic detection and evaluate its potential for predictions using in-the-wild webcam recordings. Overall, our findings demonstrate the capacity for above chance detections in both settings, with substantially better performance for the video-independent variant. Beyond this, we investigate the role of person-specific recollection biases for predictions of our video-independent models and the importance of specific modalities of facial behavior. Finally, we discuss the implications of our findings for detecting recollections and user-modeling in adaptive systems.},
    booktitle = {Proceedings of the 2022 International Conference on Multimodal Interaction},
    pages = {236–246},
    numpages = {11},
    keywords = {Memories, Affective Computing, Recollection, Cognitive Processing, User-Modeling, Mind-Wandering, Facial Behavior Analysis},
    location = {Bengaluru, India},
    series = {ICMI '22}
    }

  • E. van Krieken, E. Acar, and F. van Harmelen, “Analyzing differentiable fuzzy logic operators,” Artificial Intelligence, vol. 302, p. 103602, 2022.
    [BibTeX] [Abstract] [Download PDF]

    The AI community is increasingly putting its attention towards combining symbolic and neural approaches, as it is often argued that the strengths and weaknesses of these approaches are complementary. One recent trend in the literature are weakly supervised learning techniques that employ operators from fuzzy logics. In particular, these use prior background knowledge described in such logics to help the training of a neural network from unlabeled and noisy data. By interpreting logical symbols using neural networks, this background knowledge can be added to regular loss functions, hence making reasoning a part of learning. We study, both formally and empirically, how a large collection of logical operators from the fuzzy logic literature behave in a differentiable learning setting. We find that many of these operators, including some of the most well-known, are highly unsuitable in this setting. A further finding concerns the treatment of implication in these fuzzy logics, and shows a strong imbalance between gradients driven by the antecedent and the consequent of the implication. Furthermore, we introduce a new family of fuzzy implications (called sigmoidal implications) to tackle this phenomenon. Finally, we empirically show that it is possible to use Differentiable Fuzzy Logics for semi-supervised learning, and compare how different operators behave in practice. We find that, to achieve the largest performance improvement over a supervised baseline, we have to resort to non-standard combinations of logical operators which perform well in learning, but no longer satisfy the usual logical laws.

    @article{van2022analyzing,
    title = {Analyzing differentiable fuzzy logic operators},
    author = {van Krieken, Emile and Acar, Erman and van Harmelen,
    Frank},
    journal = {Artificial Intelligence},
    volume = {302},
    pages = {103602},
    year = {2022},
    publisher = {Elsevier},
    url = "https://research.vu.nl/ws/portalfiles/portal/146020254/2002.06100v2.pdf",
    abstract = "The AI community is increasingly putting its
    attention towards combining symbolic and neural
    approaches, as it is often argued that the strengths
    and weaknesses of these approaches are
    complementary. One recent trend in the literature
    are weakly supervised learning techniques that
    employ operators from fuzzy logics. In particular,
    these use prior background knowledge described in
    such logics to help the training of a neural network
    from unlabeled and noisy data. By interpreting
    logical symbols using neural networks, this
    background knowledge can be added to regular loss
    functions, hence making reasoning a part of
    learning. We study, both formally and empirically,
    how a large collection of logical operators from the
    fuzzy logic literature behave in a differentiable
    learning setting. We find that many of these
    operators, including some of the most well-known,
    are highly unsuitable in this setting. A further
    finding concerns the treatment of implication in
    these fuzzy logics, and shows a strong imbalance
    between gradients driven by the antecedent and the
    consequent of the implication. Furthermore, we
    introduce a new family of fuzzy implications (called
    sigmoidal implications) to tackle this
    phenomenon. Finally, we empirically show that it is
    possible to use Differentiable Fuzzy Logics for
    semi-supervised learning, and compare how different
    operators behave in practice. We find that, to
    achieve the largest performance improvement over a
    supervised baseline, we have to resort to
    non-standard combinations of logical operators which
    perform well in learning, but no longer satisfy the
    usual logical laws."
    }

  • D. W. Romero, A. Kuzina, E. J. Bekkers, J. M. Tomczak, and M. Hoogendoorn, “CKConv: Continuous Kernel Convolution For Sequential Data,” International Conference on Learning Representations (ICLR), 2022, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Conventional neural architectures for sequential data present important limitations. Recurrent networks suffer from exploding and vanishing gradients, small effective memory horizons, and must be trained sequentially. Convolutional networks are unable to handle sequences of unknown size and their memory horizon must be defined a priori. In this work, we show that all these problems can be solved by formulating convolutional kernels in CNNs as continuous functions. The resulting Continuous Kernel Convolution (CKConv) allows us to model arbitrarily long sequences in a parallel manner, within a single operation, and without relying on any form of recurrence. We show that Continuous Kernel Convolutional Networks (CKCNNs) obtain state-of-the-art results in multiple datasets, e.g., permuted MNIST, and, thanks to their continuous nature, are able to handle non-uniformly sampled datasets and irregularly-sampled data natively. CKCNNs match or perform better than neural ODEs designed for these purposes in a faster and simpler manner.

    @article{DBLP:journals/corr/abs-2102-02611,
    author = {David W. Romero and Anna Kuzina and Erik J. Bekkers
    and Jakub M. Tomczak and Mark Hoogendoorn},
    title = {CKConv: Continuous Kernel Convolution For Sequential
    Data},
    journal = {International Conference on Learning Representations
    (ICLR), 2022},
    year = {2022},
    url = {https://openreview.net/pdf?id=8FhxBtXSl0},
    abstract = {Conventional neural architectures for sequential
    data present important limitations. Recurrent
    networks suffer from exploding and vanishing
    gradients, small effective memory horizons, and must
    be trained sequentially. Convolutional networks are
    unable to handle sequences of unknown size and their
    memory horizon must be defined a priori. In this
    work, we show that all these problems can be solved
    by formulating convolutional kernels in CNNs as
    continuous functions. The resulting Continuous
    Kernel Convolution (CKConv) allows us to model
    arbitrarily long sequences in a parallel manner,
    within a single operation, and without relying on
    any form of recurrence. We show that Continuous
    Kernel Convolutional Networks (CKCNNs) obtain
    state-of-the-art results in multiple datasets, e.g.,
    permuted MNIST, and, thanks to their continuous
    nature, are able to handle non-uniformly sampled
    datasets and irregularly-sampled data
    natively. CKCNNs match or perform better than neural
    ODEs designed for these purposes in a faster and
    simpler manner.}
    }

  • F. Sarvi, M. Heuss, M. Aliannejadi, S. Schelter, and M. de Rijke, “Understanding and Mitigating the Effect of Outliers in Fair Ranking,” in WSDM 2022: The Fifteenth International Conference on Web Search and Data Mining, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Traditional ranking systems are expected to sort items in the order of their relevance and thereby maximize their utility. In fair ranking, utility is complemented with fairness as an optimization goal. Recent work on fair ranking focuses on developing algorithms to optimize for fairness, given position-based exposure. In contrast, we identify the potential of outliers in a ranking to influence exposure and thereby negatively impact fairness. An outlier in a list of items can alter the examination probabilities, which can lead to different distributions of attention, compared to position-based exposure. We formalize outlierness in a ranking, show that outliers are present in realistic datasets, and present the results of an eye-tracking study, showing that users scanning order and the exposure of items are influenced by the presence of outliers. We then introduce OMIT, a method for fair ranking in the presence of outliers. Given an outlier detection method, OMIT improves fair allocation of exposure by suppressing outliers in the top-k ranking. Using an academic search dataset, we show that outlierness optimization leads to a fairer policy that displays fewer outliers in the top-k, while maintaining a reasonable trade-off between fairness and utility.

    @inproceedings{sarvi-2022-understanding,
    author = {Sarvi, Fatemeh and Heuss, Maria and Aliannejadi,
    Mohammad and Schelter, Sebastian and de Rijke,
    Maarten},
    booktitle = {WSDM 2022: The Fifteenth International Conference on
    Web Search and Data Mining},
    month = {February},
    publisher = {ACM},
    title = {Understanding and Mitigating the Effect of Outliers
    in Fair Ranking},
    year = {2022},
    url = {https://arxiv.org/abs/2112.11251},
    abstract = "Traditional ranking systems are expected to sort
    items in the order of their relevance and thereby
    maximize their utility. In fair ranking, utility is
    complemented with fairness as an optimization
    goal. Recent work on fair ranking focuses on
    developing algorithms to optimize for fairness,
    given position-based exposure. In contrast, we
    identify the potential of outliers in a ranking to
    influence exposure and thereby negatively impact
    fairness. An outlier in a list of items can alter
    the examination probabilities, which can lead to
    different distributions of attention, compared to
    position-based exposure. We formalize outlierness in
    a ranking, show that outliers are present in
    realistic datasets, and present the results of an
    eye-tracking study, showing that users scanning
    order and the exposure of items are influenced by
    the presence of outliers. We then introduce OMIT, a
    method for fair ranking in the presence of
    outliers. Given an outlier detection method, OMIT
    improves fair allocation of exposure by suppressing
    outliers in the top-k ranking. Using an academic
    search dataset, we show that outlierness
    optimization leads to a fairer policy that displays
    fewer outliers in the top-k, while maintaining a
    reasonable trade-off between fairness and utility."
    }

  • E. Liscio, M. van der Meer, L. C. Siebert, C. M. Jonker, and P. K. Murukannaiah, “What values should an agent align with?,” Autonomous Agents and Multi-Agent Systems, vol. 36, iss. 23, p. 32, 2022.
    [BibTeX] [Abstract] [Download PDF]

    The pursuit of values drives human behavior and promotes cooperation. Existing research is focused on general values (e.g., Schwartz) that transcend contexts. However, context-specific values are necessary to (1) understand human decisions, and (2) engineer intelligent agents that can elicit and align with human values. We propose Axies, a hybrid (human and AI) methodology to identify context-specific values. Axies simplifies the abstract task of value identification as a guided value annotation process involving human annotators. Axies exploits the growing availability of value-laden text corpora and Natural Language Processing to assist the annotators in systematically identifying context-specific values. We evaluate Axies in a user study involving 80 human subjects. In our study, six annotators generate value lists for two timely and important contexts: Covid-19 measures and sustainable Energy. We employ two policy experts and 72 crowd workers to evaluate Axies value lists and compare them to a list of general (Schwartz) values. We find that Axies yields values that are (1) more context-specific than general values, (2) more suitable for value annotation than general values, and (3) independent of the people applying the methodology.

    @article{Liscio2022,
    author = {Liscio, Enrico and van der Meer, Michiel and
    Siebert, Luciano C. and Jonker, Catholijn M. and
    Murukannaiah, Pradeep K.},
    url =
    {https://link.springer.com/content/pdf/10.1007/s10458-022-09550-0},
    journal = {Autonomous Agents and Multi-Agent Systems},
    number = {23},
    pages = {32},
    publisher = {Springer US},
    title = {{What values should an agent align with?}},
    volume = {36},
    year = {2022},
    abstract = {The pursuit of values drives human behavior and
    promotes cooperation. Existing research is focused
    on general values (e.g., Schwartz) that transcend
    contexts. However, context-specific values are
    necessary to (1) understand human decisions, and (2)
    engineer intelligent agents that can elicit and
    align with human values. We propose Axies, a hybrid
    (human and AI) methodology to identify
    context-specific values. Axies simplifies the
    abstract task of value identification as a guided
    value annotation process involving human
    annotators. Axies exploits the growing availability
    of value-laden text corpora and Natural Language
    Processing to assist the annotators in
    systematically identifying context-specific
    values. We evaluate Axies in a user study involving
    80 human subjects. In our study, six annotators
    generate value lists for two timely and important
    contexts: Covid-19 measures and sustainable
    Energy. We employ two policy experts and 72 crowd
    workers to evaluate Axies value lists and compare
    them to a list of general (Schwartz) values. We find
    that Axies yields values that are (1) more
    context-specific than general values, (2) more
    suitable for value annotation than general values,
    and (3) independent of the people applying the
    methodology.}
    }

  • G. Nadizar, E. Medvet, and K. Miras, “On the Schedule for Morphological Development of Evolved Modular Soft Robots,” in European Conference on Genetic Programming (Part of EvoStar), 2022, p. 146–161.
    [BibTeX] [Abstract] [Download PDF]

    Development is fundamental for living beings. As robots are often designed to mimic biological organisms, development is believed to be crucial for achieving successful results in robotic agents, as well. What is not clear, though, is the most appropriate scheduling for development. While in real life systems development happens mostly during the initial growth phase of organisms, it has not yet been investigated whether such assumption holds also for artificial creatures. In this paper, we employ a evolutionary approach to optimize the development – according to different representations – of Voxel-based Soft Robots (VSRs), a kind of modular robots. In our study, development consists in the addition of new voxels to the VSR, at fixed time instants, depending on the development schedule. We experiment with different schedules and show that, similarly to living organisms, artificial agents benefit from development occurring at early stages of life more than from development lasting for their entire life.

    @inproceedings{nadizar2022schedule,
    title = {On the Schedule for Morphological Development of
    Evolved Modular Soft Robots},
    author = {Nadizar, Giorgia and Medvet, Eric and Miras, Karine},
    booktitle = {European Conference on Genetic Programming (Part of
    EvoStar)},
    pages = {146--161},
    year = {2022},
    organization = {Springer},
    url =
    {https://link.springer.com/chapter/10.1007/978-3-031-02056-8_10},
    abstract = {Development is fundamental for living beings. As
    robots are often designed to mimic biological
    organisms, development is believed to be crucial for
    achieving successful results in robotic agents, as
    well. What is not clear, though, is the most
    appropriate scheduling for development. While in
    real life systems development happens mostly during
    the initial growth phase of organisms, it has not
    yet been investigated whether such assumption holds
    also for artificial creatures. In this paper, we
    employ a evolutionary approach to optimize the
    development - according to different representations
    - of Voxel-based Soft Robots (VSRs), a kind of
    modular robots. In our study, development consists
    in the addition of new voxels to the VSR, at fixed
    time instants, depending on the development
    schedule. We experiment with different schedules and
    show that, similarly to living organisms, artificial
    agents benefit from development occurring at early
    stages of life more than from development lasting
    for their entire life.}
    }

  • J. Kiseleva, Z. Li, M. Aliannejadi, S. Mohanty, M. ter Hoeve, M. Burtsev, A. Skrynnik, A. Zholus, A. Panov, K. Srinet, A. Szlam, Y. Sun, K. Hofmann Marc-Alexandre Côté, A. Awadallah, L. Abdrazakov, I. Churin, P. Manggala, K. Naszadi, M. van der Meer, and T. Kim, “Interactive Grounded Language Understanding in a Collaborative Environment: IGLU 2021,” , 2022. doi:10.48550/ARXIV.2205.02388
    [BibTeX] [Abstract] [Download PDF]

    Human intelligence has the remarkable ability to quickly adapt to new tasks and environments. Starting from a very young age, humans acquire new skills and learn how to solve new tasks either by imitating the behavior of others or by following provided natural language instructions. To facilitate research in this direction, we propose \emph{IGLU: Interactive Grounded Language Understanding in a Collaborative Environment}. The primary goal of the competition is to approach the problem of how to build interactive agents that learn to solve a task while provided with grounded natural language instructions in a collaborative environment. Understanding the complexity of the challenge, we split it into sub-tasks to make it feasible for participants.

    @article{IGLU2022,
    author = {Kiseleva, Julia and Li, Ziming and Aliannejadi,
    Mohammad and Mohanty, Shrestha and ter Hoeve,
    Maartje and Burtsev, Mikhail and Skrynnik, Alexey
    and Zholus, Artem and Panov, Aleksandr and Srinet,
    Kavya and Szlam, Arthur and Sun, Yuxuan and Hofmann,
    Marc-Alexandre Côté, Katja and Awadallah, Ahmed and
    Abdrazakov, Linar and Churin, Igor and Manggala,
    Putra and Naszadi, Kata and van der Meer, Michiel
    and Kim, Taewoon},
    keywords = {Computation and Language (cs.CL), Artificial
    Intelligence (cs.AI), FOS: Computer and information
    sciences, FOS: Computer and information sciences},
    title = {Interactive Grounded Language Understanding in a
    Collaborative Environment: IGLU 2021},
    publisher = {arXiv},
    year = {2022},
    abstract = "Human intelligence has the remarkable ability to
    quickly adapt to new tasks and
    environments. Starting from a very young age, humans
    acquire new skills and learn how to solve new tasks
    either by imitating the behavior of others or by
    following provided natural language instructions. To
    facilitate research in this direction, we propose
    \emph{IGLU: Interactive Grounded Language
    Understanding in a Collaborative Environment}. The
    primary goal of the competition is to approach the
    problem of how to build interactive agents that
    learn to solve a task while provided with grounded
    natural language instructions in a collaborative
    environment. Understanding the complexity of the
    challenge, we split it into sub-tasks to make it
    feasible for participants.",
    year = {2022},
    doi = {10.48550/ARXIV.2205.02388},
    url = {https://arxiv.org/abs/2205.02388},
    copyright = {Creative Commons Attribution Share Alike 4.0
    International}
    }

  • D. Grossi, “Social Choice Around the Block: On the Computational Social Choice of Blockchain,” in 21st International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2022, Auckland, New Zealand, May 9-13, 2022, 2022, p. 1788–1793.
    [BibTeX] [Abstract] [Download PDF]

    One of the most innovative aspects of blockchain technology con- sists in the introduction of an incentive layer to regulate the behav- ior of distributed protocols. The designer of a blockchain system faces therefore issues that are akin to those relevant for the design of economic mechanisms, and faces them in a computational setting. From this perspective the present paper argues for the importance of computational social choice in blockchain research. It identifies a few challenges at the interface of the two fields that illustrate the strong potential for cross-fertilization between them.

    @inproceedings{DBLP:conf/atal/Grossi22,
    author = {Davide Grossi},
    editor = {Piotr Faliszewski and Viviana Mascardi and Catherine
    Pelachaud and Matthew E. Taylor},
    title = {Social Choice Around the Block: On the Computational
    Social Choice of Blockchain},
    booktitle = {21st International Conference on Autonomous Agents
    and Multiagent Systems, {AAMAS} 2022, Auckland, New
    Zealand, May 9-13, 2022},
    pages = {1788--1793},
    publisher = {International Foundation for Autonomous Agents and
    Multiagent Systems {(IFAAMAS)}},
    year = {2022},
    url =
    {https://www.ifaamas.org/Proceedings/aamas2022/pdfs/p1788.pdf},
    abstract = {One of the most innovative aspects of blockchain
    technology con- sists in the introduction of an
    incentive layer to regulate the behav- ior of
    distributed protocols. The designer of a blockchain
    system faces therefore issues that are akin to those
    relevant for the design of economic mechanisms, and
    faces them in a computational setting. From this
    perspective the present paper argues for the
    importance of computational social choice in
    blockchain research. It identifies a few challenges
    at the interface of the two fields that illustrate
    the strong potential for cross-fertilization between
    them.}
    }

  • M. G. Atigh, J. Schoep, E. Acar, N. van Noord, and P. Mettes, “Hyperbolic Image Segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 4453-4462.
    [BibTeX] [Abstract] [Download PDF]

    For image segmentation, the current standard is to perform pixel-level optimization and inference in Euclidean output embedding spaces through linear hyperplanes. In this work, we show that hyperbolic manifolds provide a valuable alternative for image segmentation and propose a tractable formulation of hierarchical pixel-level classification in hyperbolic space. Hyperbolic Image Segmentation opens up new possibilities and practical benefits for segmentation, such as uncertainty estimation and boundary information for free, zero-label generalization, and increased performance in low-dimensional output embeddings.

    @InProceedings{Atigh_2022_CVPR,
    author = {Atigh, Mina Ghadimi and Schoep, Julian and Acar,
    Erman and van Noord, Nanne and Mettes, Pascal},
    title = {Hyperbolic Image Segmentation},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer
    Vision and Pattern Recognition (CVPR)},
    month = {June},
    year = {2022},
    pages = {4453-4462},
    url =
    {https://openaccess.thecvf.com/content/CVPR2022/papers/Atigh_Hyperbolic_Image_Segmentation_CVPR_2022_paper.pdf},
    abstract = {For image segmentation, the current standard is to
    perform pixel-level optimization and inference in
    Euclidean output embedding spaces through linear
    hyperplanes. In this work, we show that hyperbolic
    manifolds provide a valuable alternative for image
    segmentation and propose a tractable formulation of
    hierarchical pixel-level classification in
    hyperbolic space. Hyperbolic Image Segmentation
    opens up new possibilities and practical benefits
    for segmentation, such as uncertainty estimation and
    boundary information for free, zero-label
    generalization, and increased performance in
    low-dimensional output embeddings.}
    }

  • R. Verma and E. Nalisnick, “Calibrated Learning to Defer with One-vs-All Classifiers,” in ICML 2022 Workshop on Human-Machine Collaboration and Teaming, 2022.
    [BibTeX] [Abstract] [Download PDF]

    The learning to defer (L2D) framework has the potential to make AI systems safer. For a given input, the system can defer the decision to a human if the human is more likely than the model to take the correct action. We study the calibration of L2D systems, investigating if the probabilities they output are sound. We find that Mozannar & Sontag’s (2020) multiclass framework is not calibrated with respect to expert correctness. Moreover, it is not even guaranteed to produce valid probabilities due to its parameterization being degenerate for this purpose. We propose an L2D system based on one-vs-all classifiers that is able to produce calibrated probabilities of expert correctness. Furthermore, our loss function is also a consistent surrogate for multiclass L2D, like Mozannar & Sontag’s (2020). Our experiments verify that not only is our system calibrated, but this benefit comes at no cost to accuracy. Our model’s accuracy is always comparable (and often superior) to Mozannar & Sontag’s (2020) model’s in tasks ranging from hate speech detection to galaxy classification to diagnosis of skin lesions.

    @inproceedings{Verma-Nalisnick-ICML:2022,
    title = {Calibrated Learning to Defer with One-vs-All
    Classifiers},
    author = {Rajeev Verma and Eric Nalisnick},
    year = {2022},
    booktitle = {ICML 2022 Workshop on Human-Machine Collaboration
    and Teaming},
    abstract = {The learning to defer (L2D) framework has the
    potential to make AI systems safer. For a given
    input, the system can defer the decision to a human
    if the human is more likely than the model to take
    the correct action. We study the calibration of L2D
    systems, investigating if the probabilities they
    output are sound. We find that Mozannar &
    Sontag’s (2020) multiclass
    framework is not calibrated with respect to expert
    correctness. Moreover, it is not even guaranteed to
    produce valid probabilities due to its
    parameterization being degenerate for this
    purpose. We propose an L2D system based on
    one-vs-all classifiers that is able to produce
    calibrated probabilities of expert
    correctness. Furthermore, our loss function is also
    a consistent surrogate for multiclass L2D, like
    Mozannar & Sontag's (2020). Our experiments verify
    that not only is our system calibrated, but this
    benefit comes at no cost to accuracy. Our model's
    accuracy is always comparable (and often superior)
    to Mozannar & Sontag's (2020) model's in tasks
    ranging from hate speech detection to galaxy
    classification to diagnosis of skin lesions.},
    url =
    {https://icml.cc/Conferences/2022/ScheduleMultitrack?event=18123}
    }

  • P. Manggala, H. H. Hoos, and E. Nalisnick, “Bayesian Weak Supervision via an Optimal Transport Approach,” in ICML 2022 Workshop on Human-Machine Collaboration and Teaming, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Large-scale machine learning is often impeded by a lack of labeled training data. To address this problem, the paradigm of weak supervision aims to collect and then aggregate multiple noisy labels. We propose a Bayesian probabilistic model that employs a tractable Sinkhorn-based optimal transport formulation to derive a ground-truth label. The translation between true and weak labels is cast as a transport problem with an inferred cost structure. Our approach achieves strong performance on the WRENCH weak supervision benchmark. Moreover, the posterior distribution over cost matrices allows for exploratory analysis of the weak sources.

    @inproceedings{manggala2022optimaltransportweaksupervision,
    title = {Bayesian Weak Supervision via an Optimal Transport
    Approach},
    author = {Manggala, Putra and Hoos, Holger H. and Nalisnick,
    Eric},
    year = {2022},
    booktitle = {ICML 2022 Workshop on Human-Machine Collaboration
    and Teaming},
    abstract = {Large-scale machine learning is often impeded by a
    lack of labeled training data. To address this
    problem, the paradigm of weak supervision aims to
    collect and then aggregate multiple noisy labels.
    We propose a Bayesian probabilistic model that
    employs a tractable Sinkhorn-based optimal transport
    formulation to derive a ground-truth label. The
    translation between true and weak labels is cast as
    a transport problem with an inferred cost structure.
    Our approach achieves strong performance on the
    WRENCH weak supervision benchmark. Moreover, the
    posterior distribution over cost matrices allows for
    exploratory analysis of the weak sources.},
    url = {https://openreview.net/forum?id=YJkf-6tTFiY}
    }

  • R. Dobbe, “System Safety and Artificial Intelligence,” in Oxford Handbook of AI Governance, Oxford: , 2022, vol. To Appear.
    [BibTeX] [Abstract] [Download PDF]

    This chapter formulates seven lessons for preventing harm in artificial intelligence (AI) systems based on insights from the field of system safety for software-based automation in safety-critical domains. New applications of AI across societal domains and public organizations and infrastructures come with new hazards, which lead to new forms of harm, both grave and pernicious. The text addresses the lack of consensus for diagnosing and eliminating new AI system hazards. For decades, the field of system safety has dealt with accidents and harm in safety-critical systems governed by varying degrees of software-based automation and decision-making. This field embraces the core assumption of systems and control that AI systems cannot be safeguarded by technical design choices on the model or algorithm alone, instead requiring an end-to-end hazard analysis and design frame that includes the context of use, impacted stakeholders and the formal and informal institutional environment in which the system operates. Safety and other values are then inherently socio-technical and emergent system properties that require design and control measures to instantiate these across the technical, social and institutional components of a system. This chapter honors system safety pioneer Nancy Leveson, by situating her core lessons for today’s AI system safety challenges. For every lesson, concrete tools are offered for rethinking and reorganizing the safety management of AI systems, both in design and governance. This history tells us that effective AI safety management requires transdisciplinary approaches and a shared language that allows involvement of all levels of society.

    @incollection{dobbe_system_2022,
    address = {Oxford},
    title = {System {Safety} and {Artificial} {Intelligence}},
    volume = {To Appear},
    isbn = {978-0-19-757932-9},
    url = {https://arxiv.org/abs/2202.09292},
    abstract = {This chapter formulates seven lessons for preventing
    harm in artificial intelligence (AI) systems based
    on insights from the field of system safety for
    software-based automation in safety-critical
    domains. New applications of AI across societal
    domains and public organizations and infrastructures
    come with new hazards, which lead to new forms of
    harm, both grave and pernicious. The text addresses
    the lack of consensus for diagnosing and eliminating
    new AI system hazards. For decades, the field of
    system safety has dealt with accidents and harm in
    safety-critical systems governed by varying degrees
    of software-based automation and
    decision-making. This field embraces the core
    assumption of systems and control that AI systems
    cannot be safeguarded by technical design choices on
    the model or algorithm alone, instead requiring an
    end-to-end hazard analysis and design frame that
    includes the context of use, impacted stakeholders
    and the formal and informal institutional
    environment in which the system operates. Safety and
    other values are then inherently socio-technical and
    emergent system properties that require design and
    control measures to instantiate these across the
    technical, social and institutional components of a
    system. This chapter honors system safety pioneer
    Nancy Leveson, by situating her core lessons for
    today's AI system safety challenges. For every
    lesson, concrete tools are offered for rethinking
    and reorganizing the safety management of AI
    systems, both in design and governance. This history
    tells us that effective AI safety management
    requires transdisciplinary approaches and a shared
    language that allows involvement of all levels of
    society.},
    booktitle = {Oxford {Handbook} of {AI} {Governance}},
    author = {Dobbe, Roel},
    year = {2022}
    }

  • A. Sauter, E. Acar, and V. François-Lavet, A Meta-Reinforcement Learning Algorithm for Causal DiscoveryarXiv, 2022. doi:10.48550/ARXIV.2207.08457
    [BibTeX] [Abstract] [Download PDF]

    Causal discovery is a major task with the utmost importance for machine learning since causal structures can enable models to go beyond pure correlation-based inference and significantly boost their performance. However, finding causal structures from data poses a significant challenge both in computational effort and accuracy, let alone its impossibility without interventions in general. In this paper, we develop a meta-reinforcement learning algorithm that performs causal discovery by learning to perform interventions such that it can construct an explicit causal graph. Apart from being useful for possible downstream applications, the estimated causal graph also provides an explanation for the data-generating process. In this article, we show that our algorithm estimates a good graph compared to the SOTA approaches, even in environments whose underlying causal structure is previously unseen. Further, we make an ablation study that shows how learning interventions contribute to the overall performance of our approach. We conclude that interventions indeed help boost the performance, efficiently yielding an accurate estimate of the causal structure of a possibly unseen environment.

    @misc{Sauter22MetaRL,
    doi = {10.48550/ARXIV.2207.08457},
    url = {https://arxiv.org/abs/2207.08457},
    author = {Sauter, Andreas and Acar, Erman and François-Lavet,
    Vincent},
    title = {A Meta-Reinforcement Learning Algorithm for Causal
    Discovery},
    publisher = {arXiv},
    year = {2022},
    copyright = {Creative Commons Attribution Share Alike 4.0
    International},
    abstract = { Causal discovery is a major task with the utmost
    importance for machine learning since causal
    structures can enable models to go beyond pure
    correlation-based inference and significantly boost
    their performance. However, finding causal
    structures from data poses a significant challenge
    both in computational effort and accuracy, let alone
    its impossibility without interventions in
    general. In this paper, we develop a
    meta-reinforcement learning algorithm that performs
    causal discovery by learning to perform
    interventions such that it can construct an explicit
    causal graph. Apart from being useful for possible
    downstream applications, the estimated causal graph
    also provides an explanation for the data-generating
    process. In this article, we show that our algorithm
    estimates a good graph compared to the SOTA
    approaches, even in environments whose underlying
    causal structure is previously unseen. Further, we
    make an ablation study that shows how learning
    interventions contribute to the overall performance
    of our approach. We conclude that interventions
    indeed help boost the performance, efficiently
    yielding an accurate estimate of the causal
    structure of a possibly unseen environment.}
    }

  • M. Heuss, F. Sarvi, and M. de Rijke, “Fairness of Exposure in Light of Incomplete Exposure Estimation,” in Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, New York, NY, USA, 2022, pp. 759-769.
    [BibTeX] [Abstract] [Download PDF]

    Fairness of exposure is a commonly used notion of fairness for ranking systems. It is based on the idea that all items or item groups should get exposure proportional to the merit of the item or the collective merit of the items in the group. Often, stochastic ranking policies are used to ensure fairness of exposure. Previous work unrealistically assumes that we can reliably estimate the expected exposure for all items in each ranking produced by the stochastic policy. In this work, we discuss how to approach fairness of exposure in cases where the policy contains rankings of which, due to inter-item dependencies, we cannot reliably estimate the exposure distribution. In such cases, we cannot determine whether the policy can be considered fair. Our contributions in this paper are twofold. First, we define a method called method for finding stochastic policies that avoid showing rankings with unknown exposure distribution to the user without having to compromise user utility or item fairness. Second, we extend the study of fairness of exposure to the top-k setting and also assess method in this setting. We find that method can significantly reduce the number of rankings with unknown exposure distribution without a drop in user utility or fairness compared to existing fair ranking methods, both for full-length and top-k rankings. This is an important first step in developing fair ranking methods for cases where we have incomplete knowledge about the user’s behaviour.

    @inproceedings{heuss-2022-fairness,
    author = {Heuss, Maria and Sarvi, Fatemeh and de Rijke,
    Maarten},
    title = {Fairness of Exposure in Light of Incomplete Exposure
    Estimation},
    year = {2022},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    url =
    {https://irlab.science.uva.nl/wp-content/papercite-data/pdf/heuss-2022-fairness.pdf},
    booktitle = {Proceedings of the 45th International ACM SIGIR
    Conference on Research and Development in
    Information Retrieval},
    pages = {759-769},
    location = {Madrid, Spain},
    abstract = {Fairness of exposure is a commonly used notion of
    fairness for ranking systems. It is based on the
    idea that all items or item groups should get
    exposure proportional to the merit of the item or
    the collective merit of the items in the
    group. Often, stochastic ranking policies are used
    to ensure fairness of exposure. Previous work
    unrealistically assumes that we can reliably
    estimate the expected exposure for all items in each
    ranking produced by the stochastic policy. In this
    work, we discuss how to approach fairness of
    exposure in cases where the policy contains rankings
    of which, due to inter-item dependencies, we cannot
    reliably estimate the exposure distribution. In such
    cases, we cannot determine whether the policy can be
    considered fair. Our contributions in this paper are
    twofold. First, we define a method called method for
    finding stochastic policies that avoid showing
    rankings with unknown exposure distribution to the
    user without having to compromise user utility or
    item fairness. Second, we extend the study of
    fairness of exposure to the top-k setting and also
    assess method in this setting. We find that method
    can significantly reduce the number of rankings with
    unknown exposure distribution without a drop in user
    utility or fairness compared to existing fair
    ranking methods, both for full-length and top-k
    rankings. This is an important first step in
    developing fair ranking methods for cases where we
    have incomplete knowledge about the user's
    behaviour.}
    }

  • L. Ho, V. de Boer, B. M. van Riemsdijk, S. Schlobach, and M. Tielman, “Argumentation for Knowledge Base Inconsistencies in Hybrid Intelligence Scenarios,” , online, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Hybrid Intelligence (HI) is the combination of human and machine intelligence, expanding human intellect instead of replacing it. In HI scenarios, inconsistencies in knowledge bases (KBs) can occur for a variety of reasons. These include shifting preferences, user’s motivation and or external conditions (for example, available resources and environment can vary over time). Argumentation is a potential method to address such inconsistencies as it provides a mechanism for reasoning with conflicting information, with natural explanations that are understandable to humans. In this paper, we investigate the capabilities of Argumentation in representing and reasoning about knowledge of both human and artificial agents in the presence of inconsistency. Moreover, we show how Argumentation enables Explainability for addressing problems in Decision-Making and Justification of an opinion. In order to investigate the applicability of Argumentation in HI scenarios, we demonstrate a mapping of two specific HI scenarios to Argumentation problems. We analyse to what extent of Argumentation is applicable by clarifying the practical inconsistency types of the HI scenarios that Argumentation can address. These include inconsistencies related to recommendations and decision making. We then model particularly the presentation of conflicting information for each scenario based on the form of argument representation.

    @inproceedings{HHAI2022,
    address = {online},
    author = {Ho, Loan and de Boer, Victor and van Riemsdijk,
    M. Birna and Schlobach, Stefan and Tielman, Myrthe},
    publisher = {The 1st International Workshop on Knowledge
    Representation for Hybrid Intelligence (KR4HI)},
    title = {Argumentation for Knowledge Base Inconsistencies in
    Hybrid Intelligence Scenarios},
    year = {2022},
    url =
    "https://drive.google.com/file/d/1QD95kbvzej0mXCxMQDkHOctklqCjMWO6/view",
    abstract = "Hybrid Intelligence (HI) is the combination of human
    and machine intelligence, expanding human intellect
    instead of replacing it. In HI scenarios,
    inconsistencies in knowledge bases (KBs) can occur
    for a variety of reasons. These include shifting
    preferences, user’s motivation and or external
    conditions (for example, available resources and
    environment can vary over time). Argumentation is a
    potential method to address such inconsistencies as
    it provides a mechanism for reasoning with
    conflicting information, with natural explanations
    that are understandable to humans. In this paper, we
    investigate the capabilities of Argumentation in
    representing and reasoning about knowledge of both
    human and artificial agents in the presence of
    inconsistency. Moreover, we show how Argumentation
    enables Explainability for addressing problems in
    Decision-Making and Justification of an opinion. In
    order to investigate the applicability of
    Argumentation in HI scenarios, we demonstrate a
    mapping of two specific HI scenarios to
    Argumentation problems. We analyse to what extent of
    Argumentation is applicable by clarifying the
    practical inconsistency types of the HI scenarios
    that Argumentation can address. These include
    inconsistencies related to recommendations and
    decision making. We then model particularly the
    presentation of conflicting information for each
    scenario based on the form of argument
    representation.",
    keywords = {Hybrid Intelligence, Knowledge Representation and
    Reasoning, Argumentation, Explainability,
    Inconsistency, Preferences}
    }

  • L. Ho, S. Arch-int, E. Acar, S. Schlobach, and N. Arch-int, “An argumentative approach for handling inconsistency in prioritized Datalog± ontologies,” AI Commun., vol. 35, iss. 3, pp. 243-267, 2022. doi:10.3233/AIC-220087
    [BibTeX] [Abstract] [Download PDF]

    Prioritized Datalog± is a well-studied formalism for modelling ontological knowledge and data, and has a success story in many applications in the (Semantic) Web and in other domains. Since the information content on the Web is both inherently context-dependent and frequently updated, the occurrence of a logical inconsistency is often inevitable. This phenomenon has led the research community to develop various types of inconsistency-tolerant semantics over the last few decades. Although the study of query answering under inconsistency-tolerant semantics is well-understood, the problem of explaining query answering under such semantics took considerably less attention, especially in the scenario where the facts are prioritized. In this paper, we aim to fill this gap. More specifically, we use Dung”s abstract argumentation framework to address the problem of explaining inconsistency-tolerant query answering in Datalog± KB where facts are prioritized, or preordered. We clarify the relationship between preferred repair semantics and various notions of extensions for argumentation frameworks. The strength of such argumentation-based approach is the explainability; users can more easily understand why different points of views are conflicting and why the query answer is entailed (or not) under different semantics. To this end we introduce the formal notion of a dialogical explanation, and show how it can be used to both explain showing why query results hold and not hold according to the known semantics in inconsistent Datalog± knowledge bases.

    @article{10.3233/AIC-220087,
    author = {Ho, Loan and Arch-int, Somjit and Acar, Erman and
    Schlobach, Stefan and Arch-int, Ngamnij},
    title = {An argumentative approach for handling inconsistency
    in prioritized Datalog± ontologies},
    year = {2022},
    journal = {AI Commun.},
    month = {jan},
    pages = {243-267},
    numpages = {25},
    keywords = {Argumentation, Datalog±, inconsistency, preferences,
    prioritized knowledge bases, explanation},
    issue_date = {2022},
    publisher = {IOS Press},
    address = {NLD},
    volume = {35},
    number = {3},
    issn = {0921-7126},
    url = {https://doi.org/10.3233/AIC-220087},
    doi = {10.3233/AIC-220087},
    abstract = {Prioritized Datalog± is a well-studied formalism for
    modelling ontological knowledge and data, and has a
    success story in many applications in the (Semantic)
    Web and in other domains. Since the information
    content on the Web is both inherently
    context-dependent and frequently updated, the
    occurrence of a logical inconsistency is often
    inevitable. This phenomenon has led the research
    community to develop various types of
    inconsistency-tolerant semantics over the last few
    decades. Although the study of query answering under
    inconsistency-tolerant semantics is well-understood,
    the problem of explaining query answering under such
    semantics took considerably less attention,
    especially in the scenario where the facts are
    prioritized. In this paper, we aim to fill this
    gap. More specifically, we use Dung''s abstract
    argumentation framework to address the problem of
    explaining inconsistency-tolerant query answering in
    Datalog± KB where facts are prioritized, or
    preordered. We clarify the relationship between
    preferred repair semantics and various notions of
    extensions for argumentation frameworks. The
    strength of such argumentation-based approach is the
    explainability; users can more easily understand why
    different points of views are conflicting and why
    the query answer is entailed (or not) under
    different semantics. To this end we introduce the
    formal notion of a dialogical explanation, and show
    how it can be used to both explain showing why query
    results hold and not hold according to the known
    semantics in inconsistent Datalog± knowledge bases.}
    }

  • L. Ho, V. de Boer, B. M. van Riemsdijk, S. Schlobach, and M. Tielman, “Knowledge Representation Formalisms for Hybrid Intelligence,” , online, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Hybrid Intelligence (HI) is the combination of human and machine intelligence, expanding human intellect instead of replacing it. Information in HI scenarios is often inconsistent, e.g. due to shifting preferences, user’s motivation or conflicts arising from merged data. As it provides an intuitive mechanism for reasoning with conflicting information, with natural explanations that are understandable to humans, our hypothesis is that Dung’s Abstract Argumentation (AA) is a suitable formalism for such hybrid scenarios. This paper investigates the capabilities of Argumentation in representing and reasoning in the presence of inconsistency, and its potential for intuitive explainability to link between artificial and human actors. To this end, we conduct a survey among a number of research projects of the Hybrid Intelligence Centre1 . Within these projects we analyse the applicability of argumentation with respect to various inconsistency types stemming, for instance, from commonsense reasoning, decision making, and negotiation. The results show that 14 out of the 21 projects have to deal with inconsistent information. In half of those scenarios, the knowledge models come with natural preference relations over the information. We show that Argumentation is a suitable framework to model the specific knowledge in 10 out of 14 projects, thus indicating the potential of Abstract Argumentation for transparently dealing with inconsistencies in Hybrid Intelligence systems.

    @inproceedings{ArgXAI2022,
    address = {online},
    author = {Ho, Loan and de Boer, Victor and van Riemsdijk,
    M. Birna and Schlobach, Stefan and Tielman, Myrthe},
    publisher = {1st International Workshop on Argumentation for
    eXplainable AI (ArgXAI) co-located with 9th
    International Conference on Computational Models of
    Argument (COMMA 2022)},
    title = {Knowledge Representation Formalisms for Hybrid
    Intelligence},
    volume = {Vol-3209},
    year = {2022},
    url = "http://ceur-ws.org/Vol-3209/7787.pdf",
    month = {september},
    abstract = "Hybrid Intelligence (HI) is the combination of human
    and machine intelligence, expanding human intellect
    instead of replacing it. Information in HI scenarios
    is often inconsistent, e.g. due to shifting
    preferences, user's motivation or conflicts arising
    from merged data. As it provides an intuitive
    mechanism for reasoning with conflicting
    information, with natural explanations that are
    understandable to humans, our hypothesis is that
    Dung's Abstract Argumentation (AA) is a suitable
    formalism for such hybrid scenarios. This paper
    investigates the capabilities of Argumentation in
    representing and reasoning in the presence of
    inconsistency, and its potential for intuitive
    explainability to link between artificial and human
    actors. To this end, we conduct a survey among a
    number of research projects of the Hybrid
    Intelligence Centre1 . Within these projects we
    analyse the applicability of argumentation with
    respect to various inconsistency types stemming, for
    instance, from commonsense reasoning, decision
    making, and negotiation. The results show that 14
    out of the 21 projects have to deal with
    inconsistent information. In half of those
    scenarios, the knowledge models come with natural
    preference relations over the information. We show
    that Argumentation is a suitable framework to model
    the specific knowledge in 10 out of 14 projects,
    thus indicating the potential of Abstract
    Argumentation for transparently dealing with
    inconsistencies in Hybrid Intelligence systems.",
    keywords = {Hybrid Intelligence, Argumentation, Explainability,
    Inconsistency, Preferences}
    }

  • S. Renooij, “Relevance for Robust Bayesian Network MAP-Explanations,” in Proceedings of The 11th International Conference on Probabilistic Graphical Models, 2022, p. 13–24.
    [BibTeX] [Abstract] [Download PDF]

    In the context of explainable AI, the concept of MAP-independence was recently introduced as a means for conveying the (ir)relevance of intermediate nodes for MAP computations in Bayesian networks. In this paper, we further study the concept of MAP-independence, discuss methods for finding sets of relevant nodes, and suggest ways to use these in providing users with an explanation concerning the robustness of the MAP result.

    @InProceedings{pmlr-v186-renooij22a,
    title = {Relevance for Robust Bayesian Network
    MAP-Explanations},
    author = {Renooij, Silja},
    booktitle = {Proceedings of The 11th International Conference on
    Probabilistic Graphical Models},
    pages = {13--24},
    year = {2022},
    editor = {Salmerán, Antonio and Rumí, Rafael},
    volume = {186},
    series = {Proceedings of Machine Learning Research},
    month = {05--07 Oct},
    publisher = {PMLR},
    pdf =
    {https://proceedings.mlr.press/v186/renooij22a/renooij22a.pdf},
    url = {https://proceedings.mlr.press/v186/renooij22a.html},
    abstract = {In the context of explainable AI, the concept of
    MAP-independence was recently introduced as a means
    for conveying the (ir)relevance of intermediate
    nodes for MAP computations in Bayesian networks. In
    this paper, we further study the concept of
    MAP-independence, discuss methods for finding sets
    of relevant nodes, and suggest ways to use these in
    providing users with an explanation concerning the
    robustness of the MAP result.}
    }

  • H. Prakken and R. Ratsma, “A top-level model of case-based argumentation for explanation: Formalisation and experiments,” Argument and Computation, vol. 13, pp. 159-194, 2022.
    [BibTeX] [Abstract] [Download PDF]

    This paper proposes a formal top-level model of explaining the outputs of machine-learning-based decision-making applications and evaluates it experimentally with three data sets. The model draws on AI & law research on argumentation with cases, which models how lawyers draw analogies to past cases and discuss their relevant similarities and differences in terms of relevant factors and dimensions in the problem domain. A case-based approach is natural since the input data of machine-learning applications can be seen as cases. While the approach is motivated by legal decision making, it also applies to other kinds of decision making, such as commercial decisions about loan applications or employee hiring, as long as the outcome is binary and the input conforms to this factor- or dimension format. The model is top-level in that it can be extended with more refined accounts of similarities and differences between cases. It is shown to overcome several limitations of similar argumentation-based explanation models, which only have binary features and do not represent the tendency of features towards particular outcomes. The results of the experimental evaluation studies indicate that the model may be feasible in practice, but that further development and experimentation is needed to confirm its usefulness as an explanation model. Main challenges here are selecting from a large number of possible explanations, reducing the number of features in the explanations and adding more meaningful information to them. It also remains to be investigated how suitable our approach is for explaining non-linear models.

    @ARTICLE{p+r22,
    AUTHOR = "H. Prakken and R. Ratsma",
    TITLE = "A top-level model of case-based argumentation for
    explanation: Formalisation and experiments",
    JOURNAL = "Argument and Computation",
    YEAR = "2022",
    VOLUME = "13",
    PAGES = "159-194",
    Abstract = "This paper proposes a formal top-level model of
    explaining the outputs of machine-learning-based
    decision-making applications and evaluates it
    experimentally with three data sets. The model draws
    on AI & law research on argumentation with cases,
    which models how lawyers draw analogies to past
    cases and discuss their relevant similarities and
    differences in terms of relevant factors and
    dimensions in the problem domain. A case-based
    approach is natural since the input data of
    machine-learning applications can be seen as
    cases. While the approach is motivated by legal
    decision making, it also applies to other kinds of
    decision making, such as commercial decisions about
    loan applications or employee hiring, as long as the
    outcome is binary and the input conforms to this
    factor- or dimension format. The model is top-level
    in that it can be extended with more refined
    accounts of similarities and differences between
    cases. It is shown to overcome several limitations
    of similar argumentation-based explanation models,
    which only have binary features and do not represent
    the tendency of features towards particular
    outcomes. The results of the experimental evaluation
    studies indicate that the model may be feasible in
    practice, but that further development and
    experimentation is needed to confirm its usefulness
    as an explanation model. Main challenges here are
    selecting from a large number of possible
    explanations, reducing the number of features in the
    explanations and adding more meaningful information
    to them. It also remains to be investigated how
    suitable our approach is for explaining non-linear
    models.",
    URL =
    "https://content.iospress.com/articles/argument-and-computation/aac210009",
    }

  • M. M. Çelikok, F. A. Oliehoek, and S. Kaski, “Best-Response Bayesian Reinforcement Learning with Bayes-Adaptive POMDPs for Centaurs,” in Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems, Richland, SC, 2022, pp. 235-243.
    [BibTeX] [Abstract] [Download PDF]

    Centaurs are half-human, half-AI decision-makers where the AI’s goal is to complement the human. To do so, the AI must be able to recognize the goals and constraints of the human and have the means to help them. We present a novel formulation of the interaction between the human and the AI as a sequential game where the agents are modelled using Bayesian best-response models. We show that in this case the AI’s problem of helping bounded-rational humans make better decisions reduces to a Bayes-adaptive POMDP. In our simulated experiments, we consider an instantiation of our framework for humans who are subjectively optimistic about the AI’s future behaviour. Our results show that when equipped with a model of the human, the AI can infer the human’s bounds and nudge them towards better decisions. We discuss ways in which the machine can learn to improve upon its own limitations as well with the help of the human. We identify a novel trade-off for centaurs in partially observable tasks: for the AI’s actions to be acceptable to the human, the machine must make sure their beliefs are sufficiently aligned, but aligning beliefs might be costly. We present a preliminary theoretical analysis of this trade-off and its dependence on task structure.

    @inproceedings{10.5555/3535850.3535878,
    author = {Çelikok, Mustafa Mert and Oliehoek, Frans A. and
    Kaski, Samuel},
    title = {Best-Response Bayesian Reinforcement Learning with
    Bayes-Adaptive POMDPs for Centaurs},
    year = {2022},
    isbn = {9781450392136},
    publisher = {International Foundation for Autonomous Agents and
    Multiagent Systems},
    address = {Richland, SC},
    booktitle = {Proceedings of the 21st International Conference on
    Autonomous Agents and Multiagent Systems},
    pages = {235-243},
    numpages = {9},
    keywords = {Bayesian reinforcement learning, multiagent
    learning, computational rationality, hybrid
    intelligence},
    location = {Virtual Event, New Zealand},
    series = {AAMAS '22},
    url =
    {"https://dl.acm.org/doi/abs/10.5555/3535850.3535878"},
    abstract = {Centaurs are half-human, half-AI decision-makers
    where the AI's goal is to complement the human. To
    do so, the AI must be able to recognize the goals
    and constraints of the human and have the means to
    help them. We present a novel formulation of the
    interaction between the human and the AI as a
    sequential game where the agents are modelled using
    Bayesian best-response models. We show that in this
    case the AI's problem of helping bounded-rational
    humans make better decisions reduces to a
    Bayes-adaptive POMDP. In our simulated experiments,
    we consider an instantiation of our framework for
    humans who are subjectively optimistic about the
    AI's future behaviour. Our results show that when
    equipped with a model of the human, the AI can infer
    the human's bounds and nudge them towards better
    decisions. We discuss ways in which the machine can
    learn to improve upon its own limitations as well
    with the help of the human. We identify a novel
    trade-off for centaurs in partially observable
    tasks: for the AI's actions to be acceptable to the
    human, the machine must make sure their beliefs are
    sufficiently aligned, but aligning beliefs might be
    costly. We present a preliminary theoretical
    analysis of this trade-off and its dependence on
    task structure.}
    }

  • H. Prakken, “Formalising an aspect of argument strength: degrees of attackability,” in Computational Models of Argument. Proceedings of COMMA 2022, F. T. et al., Ed., Amsterdam etc: IOS Press, 2022, pp. 296-307. doi:10.3233/FAIA220161
    [BibTeX] [Abstract] [Download PDF]

    This paper formally studies a notion of dialectical argument strength in terms of the number of ways in which an argument can be successfully attacked in expansions of an abstract argumentation framework. The proposed model is abstract but its design is motivated by the wish to avoid overly limiting assumptions that may not hold in particular dialogue contexts or in particular structured accounts of argumentation. It is shown that most principles for gradual argument acceptability proposed in the literature fail to hold for the proposed notion of dialectical strength, which clarifies their rational foundations and highlights the importance of distinguishing between logical, dialectical and rhetorical argument strength.

    @INCOLLECTION{hp22gradual,
    AUTHOR = "H. Prakken",
    TITLE = "Formalising an aspect of argument strength: degrees
    of attackability",
    BOOKTITLE = "Computational Models of Argument. Proceedings of
    {COMMA} 2022",
    EDITOR = "Francesca Toni et al.",
    PUBLISHER = "IOS Press",
    ADDRESS = "Amsterdam etc",
    PAGES = "296-307",
    YEAR = "2022",
    abstract = "This paper formally studies a notion of dialectical
    argument strength in terms of the number of ways in
    which an argument can be successfully attacked in
    expansions of an abstract argumentation
    framework. The proposed model is abstract but its
    design is motivated by the wish to avoid overly
    limiting assumptions that may not hold in particular
    dialogue contexts or in particular structured
    accounts of argumentation. It is shown that most
    principles for gradual argument acceptability
    proposed in the literature fail to hold for the
    proposed notion of dialectical strength, which
    clarifies their rational foundations and highlights
    the importance of distinguishing between logical,
    dialectical and rhetorical argument strength.",
    DOI = "10.3233/FAIA220161",
    url = "https://ebooks.iospress.nl/doi/10.3233/FAIA220161"
    }

  • W. van Woerkom, D. Grossi, H. Prakken, and B. Verheij, “Justification in Case-Based Reasoning,” in Proceedings of the First International Workshop on Argumentation for eXplainable AI, 2022, pp. 1-13.
    [BibTeX] [Abstract] [Download PDF]

    The explanation and justification of decisions is an important subject in contemporary data-driven automated methods. Case-based argumentation has been proposed as the formal background for the explanation of data-driven automated decision making. In particular, a method was developed in recent work based on the theory of precedential constraint which reasons from a case base, given by the training data of the machine learning system, to produce a justification for the outcome of a focus case. An important role is played in this method by the notions of citability and compensation, and in the present work we develop these in more detail. Special attention is paid to the notion of compensation; we formally specify the notion and identify several of its desirable properties. These considerations reveal a refined formal perspective on the explanation method as an extension of the theory of precedential constraint with a formal notion of justification.

    @inproceedings{vanwoerkom2022,
    author = "van Woerkom, Wijnand and Grossi, Davide and Prakken,
    Henry and Verheij, Bart",
    editor = "Čyras, Kristijonas and Kampik, Timotheus and
    Cocarascu, Oana and Rago, Antonio",
    title = "Justification in Case-Based Reasoning",
    booktitle = "Proceedings of the First International Workshop on
    Argumentation for eXplainable AI",
    year = "2022",
    pages = {1-13},
    publisher = "CEUR Workshop Proceedings",
    Abstract = "The explanation and justification of decisions is an
    important subject in contemporary data-driven
    automated methods. Case-based argumentation has been
    proposed as the formal background for the
    explanation of data-driven automated decision
    making. In particular, a method was developed in
    recent work based on the theory of precedential
    constraint which reasons from a case base, given by
    the training data of the machine learning system, to
    produce a justification for the outcome of a focus
    case. An important role is played in this method by
    the notions of citability and compensation, and in
    the present work we develop these in more
    detail. Special attention is paid to the notion of
    compensation; we formally specify the notion and
    identify several of its desirable properties. These
    considerations reveal a refined formal perspective
    on the explanation method as an extension of the
    theory of precedential constraint with a formal
    notion of justification.",
    URL = "http://ceur-ws.org/Vol-3209/",
    }

  • N. Hopner, I. Tiddi, and H. van Hoof, “Leveraging Class Abstraction for Commonsense Reinforcement Learning via Residual Policy Gradient Methods,” in Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22, 2022, p. 3050–3056. doi:10.24963/ijcai.2022/423
    [BibTeX] [Abstract] [Download PDF]

    Enabling reinforcement learning (RL) agents to leverage a knowledge base while learning from experience promises to advance RL in knowledge intensive domains. However, it has proven difficult to leverage knowledge that is not manually tailored to the environment. We propose to use the subclass relationships present in open-source knowledge graphs to abstract away from specific objects. We develop a residual policy gradient method that is able to integrate knowledge across different abstraction levels in the class hierarchy. Our method results in improved sample efficiency and generalisation to unseen objects in commonsense games, but we also investigate failure modes, such as excessive noise in the extracted class knowledge or environments with little class structure.

    @inproceedings{ijcai2022p423,
    title = {Leveraging Class Abstraction for Commonsense
    Reinforcement Learning via Residual Policy Gradient
    Methods},
    author = {Hopner, Niklas and Tiddi, Ilaria and van Hoof,
    Herke},
    booktitle = {Proceedings of the Thirty-First International Joint
    Conference on Artificial Intelligence, {IJCAI-22}},
    publisher = {International Joint Conferences on Artificial
    Intelligence Organization},
    editor = {Lud De Raedt},
    pages = {3050--3056},
    year = {2022},
    month = {7},
    note = {Main Track},
    doi = {10.24963/ijcai.2022/423},
    url = {https://doi.org/10.24963/ijcai.2022/423},
    abstract = {Enabling reinforcement learning (RL) agents to
    leverage a knowledge base while learning from
    experience promises to advance RL in knowledge
    intensive domains. However, it has proven difficult
    to leverage knowledge that is not manually tailored
    to the environment. We propose to use the subclass
    relationships present in open-source knowledge
    graphs to abstract away from specific objects. We
    develop a residual policy gradient method that is
    able to integrate knowledge across different
    abstraction levels in the class hierarchy. Our
    method results in improved sample efficiency and
    generalisation to unseen objects in commonsense
    games, but we also investigate failure modes, such
    as excessive noise in the extracted class knowledge
    or environments with little class structure.}
    }

  • S. Baez Santamaria, P. Vossen, and T. Baier, “Evaluating Agent Interactions Through Episodic Knowledge Graphs,” in Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge, Gyeongju, Republic of Korea, 2022, p. 15–28.
    [BibTeX] [Abstract] [Download PDF]

    We present a new method based on episodic Knowledge Graphs (eKGs) for evaluating (multimodal) conversational agents in open domains. This graph is generated by interpreting raw signals during conversation and is able to capture the accumulation of knowledge over time. We apply structural and semantic analysis of the resulting graphs and translate the properties into qualitative measures. We compare these measures with existing automatic and manual evaluation metrics commonly used for conversational agents. Our results show that our Knowledge-Graph-based evaluation provides more qualitative insights into interaction and the agent{‘}s behavior.

    @inproceedings{baez-santamaria-etal-2022-evaluating,
    title = "Evaluating Agent Interactions Through Episodic
    Knowledge Graphs",
    author = "Baez Santamaria, Selene and Vossen, Piek and Baier,
    Thomas",
    booktitle = "Proceedings of the 1st Workshop on Customized Chat
    Grounding Persona and Knowledge",
    month = "oct",
    year = "2022",
    address = "Gyeongju, Republic of Korea",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2022.ccgpk-1.3",
    pages = "15--28",
    abstract = "We present a new method based on episodic Knowledge
    Graphs (eKGs) for evaluating (multimodal)
    conversational agents in open domains. This graph is
    generated by interpreting raw signals during
    conversation and is able to capture the accumulation
    of knowledge over time. We apply structural and
    semantic analysis of the resulting graphs and
    translate the properties into qualitative
    measures. We compare these measures with existing
    automatic and manual evaluation metrics commonly
    used for conversational agents. Our results show
    that our Knowledge-Graph-based evaluation provides
    more qualitative insights into interaction and the
    agent{'}s behavior."
    }

  • M. van der Meer, M. Reuver, U. Khurana, L. Krause, and S. Santamaría, Will It Blend? Mixing Training Paradigms & Prompting for Argument Quality PredictionarXiv, 2022. doi:10.48550/ARXIV.2209.08966
    [BibTeX] [Abstract] [Download PDF]

    This paper describes our winning contribution to the Shared Task of the 9th Workshop on Argument Mining (2022). Our approach uses Large Language Models for the task of Argument Quality Prediction. We perform prompt engineering using GPT-3, and also investigate the training paradigms multi-task learning, contrastive learning, and intermediate-task training. We find that a mixed prediction setup outperforms single models. Prompting GPT-3 works best for predicting argument validity, and argument novelty is best estimated by a model trained using all three training paradigms.

    @misc{https://doi.org/10.48550/arxiv.2209.08966,
    doi = {10.48550/ARXIV.2209.08966},
    url = {https://arxiv.org/abs/2209.08966},
    author = {van der Meer, Michiel and Reuver, Myrthe and
    Khurana, Urja and Krause, Lea and Santamaría, Selene
    Báez},
    title = {Will It Blend? Mixing Training Paradigms &
    Prompting for Argument Quality Prediction},
    publisher = {arXiv},
    year = {2022},
    abstract = {This paper describes our winning contribution to the
    Shared Task of the 9th Workshop on Argument Mining
    (2022). Our approach uses Large Language Models for
    the task of Argument Quality Prediction. We perform
    prompt engineering using GPT-3, and also investigate
    the training paradigms multi-task learning,
    contrastive learning, and intermediate-task
    training. We find that a mixed prediction setup
    outperforms single models. Prompting GPT-3 works
    best for predicting argument validity, and argument
    novelty is best estimated by a model trained using
    all three training paradigms.},
    copyright = {Creative Commons Attribution Non Commercial Share
    Alike 4.0 International}
    }

  • K. Miras and A. Eiben, “How the History of Changing Environments Affects Traits of Evolvable Robot Populations,” Artificial life, vol. 28, iss. 2, p. 224–239, 2022.
    [BibTeX] [Abstract] [Download PDF]

    The environment is one of the key factors in the emergence of intelligent creatures, but it has received little attention within the Evolutionary Robotics literature. This article investigates the effects of changing environments on morphological and behavioral traits of evolvable robots. In particular, we extend a previous study by evolving robot populations under diverse changing-environment setups, varying the magnitude, frequency, duration, and dynamics of the changes. The results show that long-lasting effects of early generations occur not only when transitioning from easy to hard conditions, but also when going from hard to easy conditions. Furthermore, we demonstrate how the impact of environmental scaffolding is dependent on the nature of the environmental changes involved.

    @article{miras2022history,
    title = {How the History of Changing Environments Affects
    Traits of Evolvable Robot Populations},
    author = {Miras, Karine and Eiben, AE},
    journal = {Artificial life},
    volume = {28},
    url = {https://research.vu.nl/en/publications/how-the-history-of-changing-environments-affects-traits-of-evolva},
    number = {2},
    pages = {224--239},
    year = {2022},
    publisher = {MIT Press One Broadway, 12th Floor, Cambridge,
    Massachusetts 02142, USA~…},
    abstract = {The environment is one of the key factors in the
    emergence of intelligent creatures, but it has
    received little attention within the Evolutionary
    Robotics literature. This article investigates the
    effects of changing environments on morphological
    and behavioral traits of evolvable robots. In
    particular, we extend a previous study by evolving
    robot populations under diverse changing-environment
    setups, varying the magnitude, frequency, duration,
    and dynamics of the changes. The results show that
    long-lasting effects of early generations occur not
    only when transitioning from easy to hard
    conditions, but also when going from hard to easy
    conditions. Furthermore, we demonstrate how the
    impact of environmental scaffolding is dependent on
    the nature of the environmental changes involved.}
    }

  • J. Schoemaker and K. Miras, “The benefits of credit assignment in noisy video game environments,” in Artificial Life Conference Proceedings 34, 2022, p. 6.
    [BibTeX] [Abstract] [Download PDF]

    Both Evolutionary Algorithms (EAs) and Reinforcement Learning Algorithms (RLAs) have proven successful in policy optimisation tasks, however, there is scarce literature comparing their strengths and weaknesses. This makes it difficult to determine which group of algorithms is best suited for a task. This paper presents a comparison of two EAs and two RLAs in solving EvoMan – a video game playing benchmark. We test the algorithms both with and without noise introduction in the initialisation of multiple video game environments. We demonstrate that EAs reach a similar performance to RLAs in the static environments, but when noise is introduced the performance of EAs drops drastically while the performance of RLAs is much less affected.

    @inproceedings{schoemaker2022benefits,
    title = {The benefits of credit assignment in noisy video
    game environments},
    author = {Schoemaker, Jacob and Miras, Karine},
    booktitle = {Artificial Life Conference Proceedings 34},
    volume = {2022},
    number = {1},
    pages = {6},
    url =
    {https://direct.mit.edu/isal/proceedings-pdf/isal/34/6/2035319/isal_a_00483.pdf},
    year = {2022},
    organization = {MIT Press One Rogers Street, Cambridge, MA
    02142-1209, USA journals-info~…},
    abstract = {Both Evolutionary Algorithms (EAs) and Reinforcement
    Learning Algorithms (RLAs) have proven successful in
    policy optimisation tasks, however, there is scarce
    literature comparing their strengths and
    weaknesses. This makes it difficult to determine
    which group of algorithms is best suited for a
    task. This paper presents a comparison of two EAs
    and two RLAs in solving EvoMan - a video game
    playing benchmark. We test the algorithms both with
    and without noise introduction in the initialisation
    of multiple video game environments. We demonstrate
    that EAs reach a similar performance to RLAs in the
    static environments, but when noise is introduced
    the performance of EAs drops drastically while the
    performance of RLAs is much less affected.}
    }

  • W. van Woerkom, D. Grossi, H. Prakken, and B. Verheij, “Landmarks in Case-Based Reasoning: From Theory to Data,” in HHAI2022: Augmenting Human Intellect, 2022, pp. 212-224.
    [BibTeX] [Abstract] [Download PDF]

    Widespread application of uninterpretable machine learning systems for sensitive purposes has spurred research into elucidating the decision making process of these systems. These efforts have their background in many different disciplines, one of which is the field of AI & law. In particular, recent works have observed that machine learning training data can be interpreted as legal cases. Under this interpretation the formalism developed to study case law, called the theory of precedential constraint, can be used to analyze the way in which machine learning systems draw on training data – or should draw on them – to make decisions. These works predominantly stay on the theoretical level, hence in the present work the formalism is evaluated on a real world dataset. Through this analysis we identify a significant new concept which we call landmark cases, and use it to characterize the types of datasets that are more or less suitable to be described by the theory.

    @InProceedings{woerkom2022landmarks,
    author = {van Woerkom, Wijnand and Grossi, Davide and Prakken,
    Henry and Verheij, Bart},
    title = {Landmarks in Case-Based Reasoning: From Theory to
    Data},
    booktitle = {HHAI2022: Augmenting Human Intellect},
    year = {2022},
    editor = {Schlobach, Stefan and Pérez-Ortiz, María and
    Tielman, Myrthe},
    volume = {354},
    series = {Frontiers in Artificial Intelligence and
    Applications},
    pages = {212-224},
    publisher = {IOS Press},
    Abstract = {Widespread application of uninterpretable machine
    learning systems for sensitive purposes has spurred
    research into elucidating the decision making
    process of these systems. These efforts have their
    background in many different disciplines, one of
    which is the field of AI & law. In particular,
    recent works have observed that machine learning
    training data can be interpreted as legal
    cases. Under this interpretation the formalism
    developed to study case law, called the theory of
    precedential constraint, can be used to analyze the
    way in which machine learning systems draw on
    training data – or should draw on them – to make
    decisions. These works predominantly stay on the
    theoretical level, hence in the present work the
    formalism is evaluated on a real world
    dataset. Through this analysis we identify a
    significant new concept which we call landmark
    cases, and use it to characterize the types of
    datasets that are more or less suitable to be
    described by the theory.},
    URL = {https://ebooks.iospress.nl/volumearticle/60868}
    }

  • N. Kökciyan and P. Yolum, “Taking Situation-Based Privacy Decisions: Privacy Assistants Working with Humans,” in Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence (IJCAI), 2022, p. 703–709.
    [BibTeX] [Abstract] [Download PDF]

    Privacy on the Web is typically managed by giving consent to individual Websites for various aspects of data usage. This paradigm requires too much human effort and thus is impractical for Internet of Things (IoT) applications where humans interact with many new devices on a daily basis. Ideally, software privacy assistants can help by making privacy decisions in different situations on behalf of the users. To realize this, we propose an agent-based model for a privacy assistant. The model identifies the contexts that a situation implies and computes the trustworthiness of these contexts. Contrary to traditional trust models that capture trust in an entity by observing large number of interactions, our proposed model can assess the trustworthiness even if the user has not interacted with the particular device before. Moreover, our model can decide which situations are inherently ambiguous and thus can request the human to make the decision. We evaluate various aspects of the model using a real-life data set and report adjustments that are needed to serve different types of users well.

    @inproceedings{pas-ijcai-2022,
    title = {Taking Situation-Based Privacy Decisions: Privacy
    Assistants Working with Humans},
    author = {Kökciyan, Nadin and Yolum, Pinar},
    booktitle = {Proceedings of the Thirty-First International Joint
    Conference on Artificial Intelligence (IJCAI)},
    pages = {703--709},
    year = {2022},
    month = {7},
    abstract = {Privacy on the Web is typically managed by giving
    consent to individual Websites for various aspects
    of data usage. This paradigm requires too much human
    effort and thus is impractical for Internet of
    Things (IoT) applications where humans interact with
    many new devices on a daily basis. Ideally, software
    privacy assistants can help by making privacy
    decisions in different situations on behalf of the
    users. To realize this, we propose an agent-based
    model for a privacy assistant. The model identifies
    the contexts that a situation implies and computes
    the trustworthiness of these contexts. Contrary to
    traditional trust models that capture trust in an
    entity by observing large number of interactions,
    our proposed model can assess the trustworthiness
    even if the user has not interacted with the
    particular device before. Moreover, our model can
    decide which situations are inherently ambiguous and
    thus can request the human to make the decision. We
    evaluate various aspects of the model using a
    real-life data set and report adjustments that are
    needed to serve different types of users well.},
    url = {https://www.ijcai.org/proceedings/2022/0099.pdf}
    }

  • O. Ulusoy and P. Yolum, “PANOLA: A Personal Assistant for Supporting Users in Preserving Privacy,” ACM Transactions on Internet Technology, vol. 22, iss. 1, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Privacy is the right of individuals to keep personal information to themselves. When individuals use online systems, they should be given the right to decide what information they would like to share and what to keep private. When a piece of information pertains only to a single individual, preserving privacy is possible by providing the right access options to the user. However, when a piece of information pertains to multiple individuals, such as a picture of a group of friends or a collaboratively edited document, deciding how to share this information and with whom is challenging. The problem becomes more difficult when the individuals who are affected by the information have different, possibly conflicting privacy constraints. Resolving this problem requires a mechanism that takes into account the relevant individuals’ concerns to decide on the privacy configuration of information. Because these decisions need to be made frequently (i.e., per each piece of shared content), the mechanism should be automated. This article presents a personal assistant to help end-users with managing the privacy of their content. When some content that belongs to multiple users is about to be shared, the personal assistants of the users employ an auction-based privacy mechanism to regulate the privacy of the content. To do so, each personal assistant learns the preferences of its user over time and produces bids accordingly. Our proposed personal assistant is capable of assisting users with different personas and thus ensures that people benefit from it as they need it. Our evaluations over multiagent simulations with online social network content show that our proposed personal assistant enables privacy-respecting content sharing.

    @article{panola-2022,
    author = {Ulusoy, Onuralp and Yolum, Pinar},
    title = {{PANOLA}: A Personal Assistant for Supporting Users
    in Preserving Privacy},
    year = {2022},
    issue_date = {February 2022},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    volume = {22},
    number = {1},
    journal = {ACM Transactions on Internet Technology},
    month = {February},
    articleno = {27},
    numpages = {32},
    url = {https://doi.org/10.1145/3471187},
    abstract = {Privacy is the right of individuals to keep personal
    information to themselves. When individuals use
    online systems, they should be given the right to
    decide what information they would like to share and
    what to keep private. When a piece of information
    pertains only to a single individual, preserving
    privacy is possible by providing the right access
    options to the user. However, when a piece of
    information pertains to multiple individuals, such
    as a picture of a group of friends or a
    collaboratively edited document, deciding how to
    share this information and with whom is
    challenging. The problem becomes more difficult when
    the individuals who are affected by the information
    have different, possibly conflicting privacy
    constraints. Resolving this problem requires a
    mechanism that takes into account the relevant
    individuals’ concerns to decide on the privacy
    configuration of information. Because these
    decisions need to be made frequently (i.e., per each
    piece of shared content), the mechanism should be
    automated. This article presents a personal
    assistant to help end-users with managing the
    privacy of their content. When some content that
    belongs to multiple users is about to be shared, the
    personal assistants of the users employ an
    auction-based privacy mechanism to regulate the
    privacy of the content. To do so, each personal
    assistant learns the preferences of its user over
    time and produces bids accordingly. Our proposed
    personal assistant is capable of assisting users
    with different personas and thus ensures that people
    benefit from it as they need it. Our evaluations
    over multiagent simulations with online social
    network content show that our proposed personal
    assistant enables privacy-respecting content
    sharing.}
    }

  • Gönül. Ayci, M. Şensoy, A. Özgür, and P. Yolum, “Uncertainty-Aware Personal Assistant for Making Personalized Privacy Decisions,” ACM Transactions on Internet Technology, 2022. doi:10.1145/3561820
    [BibTeX] [Abstract] [Download PDF]

    Many software systems, such as online social networks enable users to share information about themselves. While the action of sharing is simple, it requires an elaborate thought process on privacy: what to share, with whom to share, and for what purposes. Thinking about these for each piece of content to be shared is tedious. Recent approaches to tackle this problem build personal assistants that can help users by learning what is private over time and recommending privacy labels such as private or public to individual content that a user considers sharing. However, privacy is inherently ambiguous and highly personal. Existing approaches to recommend privacy decisions do not address these aspects of privacy sufficiently. Ideally, a personal assistant should be able to adjust its recommendation based on a given user, considering that user’s privacy understanding. Moreover, the personal assistant should be able to assess when its recommendation would be uncertain and let the user make the decision on her own. Accordingly, this paper proposes a personal assistant that uses evidential deep learning to classify content based on its privacy label. An important characteristic of the personal assistant is that it can model its uncertainty in its decisions explicitly, determine that it does not know the answer, and delegate from making a recommendation when its uncertainty is high. By factoring in user’s own understanding of privacy, such as risk factors or own labels, the personal assistant can personalize its recommendations per user. We evaluate our proposed personal assistant using a well-known data set. Our results show that our personal assistant can accurately identify uncertain cases, personalize them to its user’s needs, and thus helps users preserve their privacy well.

    @article{pure-2022,
    author = {Ayci, Gönül and Şensoy, Murat and Özgür, Arzucan and
    Yolum, Pinar},
    title = {Uncertainty-Aware Personal Assistant for Making
    Personalized Privacy Decisions},
    year = {2022},
    journal = {ACM Transactions on Internet Technology},
    publisher = {Association for Computing Machinery},
    url = {https://doi.org/10.1145/3561820},
    doi = {10.1145/3561820},
    abstract = {Many software systems, such as online social
    networks enable users to share information about
    themselves. While the action of sharing is simple,
    it requires an elaborate thought process on privacy:
    what to share, with whom to share, and for what
    purposes. Thinking about these for each piece of
    content to be shared is tedious. Recent approaches
    to tackle this problem build personal assistants
    that can help users by learning what is private over
    time and recommending privacy labels such as private
    or public to individual content that a user
    considers sharing. However, privacy is inherently
    ambiguous and highly personal. Existing approaches
    to recommend privacy decisions do not address these
    aspects of privacy sufficiently. Ideally, a personal
    assistant should be able to adjust its
    recommendation based on a given user, considering
    that user’s privacy understanding. Moreover, the
    personal assistant should be able to assess when its
    recommendation would be uncertain and let the user
    make the decision on her own. Accordingly, this
    paper proposes a personal assistant that uses
    evidential deep learning to classify content based
    on its privacy label. An important characteristic of
    the personal assistant is that it can model its
    uncertainty in its decisions explicitly, determine
    that it does not know the answer, and delegate from
    making a recommendation when its uncertainty is
    high. By factoring in user’s own understanding of
    privacy, such as risk factors or own labels, the
    personal assistant can personalize its
    recommendations per user. We evaluate our proposed
    personal assistant using a well-known data set. Our
    results show that our personal assistant can
    accurately identify uncertain cases, personalize
    them to its user’s needs, and thus helps users
    preserve their privacy well.},
    note = {In press}
    }

  • E. Erdogan, F. Dignum, R. Verbrugge, and P. Yolum, “Abstracting Minds: Computational Theory of Mind for Human-Agent Collaboration,” in HHAI2022: Augmenting Human Intellect, IOS Press, 2022, p. 199–211.
    [BibTeX] [Abstract] [Download PDF]

    Theory of mind refers to the human ability to reason about mental content of other people such as beliefs, desires, and goals. In everyday life, people rely on their theory of mind to understand, explain, and predict the behaviour of others. Having a theory of mind is especially useful when people collaborate, since individuals can then reason on what the other individual knows as well as what reasoning they might do. Realization of hybrid intelligence, where an agent collaborates with a human, will require the agent to be able to do similar reasoning through computational theory of mind. Accordingly, this paper provides a mechanism for computational theory of mind based on abstractions of single beliefs into higher-level concepts. These concepts can correspond to social norms, roles, as well as values. Their use in decision making serves as a heuristic to choose among interactions, thus facilitating collaboration on decisions. Using examples from the medical domain, we demonstrate how having such a theory of mind enables an agent to interact with humans efficiently and can increase the quality of the decisions humans make.

    @incollection{erdogan2022abstracting,
    title = {Abstracting Minds: Computational Theory of Mind for
    Human-Agent Collaboration},
    author = {Erdogan, Emre and Dignum, Frank and Verbrugge,
    Rineke and Yolum, Pinar},
    booktitle = {HHAI2022: Augmenting Human Intellect},
    pages = {199--211},
    year = {2022},
    publisher = {IOS Press},
    url = {http://dx.doi.org/10.3233/FAIA220199},
    abstract = {Theory of mind refers to the human ability to reason
    about mental content of other people such as
    beliefs, desires, and goals. In everyday life,
    people rely on their theory of mind to understand,
    explain, and predict the behaviour of others. Having
    a theory of mind is especially useful when people
    collaborate, since individuals can then reason on
    what the other individual knows as well as what
    reasoning they might do. Realization of hybrid
    intelligence, where an agent collaborates with a
    human, will require the agent to be able to do
    similar reasoning through computational theory of
    mind. Accordingly, this paper provides a mechanism
    for computational theory of mind based on
    abstractions of single beliefs into higher-level
    concepts. These concepts can correspond to social
    norms, roles, as well as values. Their use in
    decision making serves as a heuristic to choose
    among interactions, thus facilitating collaboration
    on decisions. Using examples from the medical
    domain, we demonstrate how having such a theory of
    mind enables an agent to interact with humans
    efficiently and can increase the quality of the
    decisions humans make.}
    }

  • E. Erdogan, F. Dignum, R. Verbrugge, and P. Yolum, “Computational Theory of Mind for Human-Agent Coordination,” in Coordination, Organizations, Institutions, Norms, and Ethics for Governance of Multi-Agent Systems XV, 2022, p. 92–108.
    [BibTeX] [Abstract] [Download PDF]

    In everyday life, people often depend on their theory of mind, i.e., their ability to reason about unobservable mental content of others to understand, explain, and predict their behaviour. Many agent-based models have been designed to develop computational theory of mind and analyze its effectiveness in various tasks and settings. However, most existing models are not generic (e.g., only applied in a given setting), not feasible (e.g., require too much information to be processed), or not human-inspired (e.g., do not capture the behavioral heuristics of humans). This hinders their applicability in many settings. Accordingly, we propose a new computational theory of mind, which captures the human decision heuristics of reasoning by abstracting individual beliefs about others. We specifically study computational affinity and show how it can be used in tandem with theory of mind reasoning when designing agent models for human-agent negotiation. We perform two-agent simulations to analyze the role of affinity in getting to agreements when there is a bound on the time to be spent for negotiating. Our results suggest that modeling affinity can ease the negotiation process by decreasing the number of rounds needed for an agreement as well as yield a higher benefit for agents with theory of mind reasoning.

    @InProceedings{erdogan+2022,
    author = "Erdogan, Emre and Dignum, Frank and Verbrugge,
    Rineke and Yolum, Pinar",
    editor = "Ajmeri, Nirav and Morris Martin, Andreasa and
    Savarimuthu, Bastin Tony Roy",
    title = "Computational Theory of Mind for Human-Agent
    Coordination",
    booktitle = "Coordination, Organizations, Institutions, Norms,
    and Ethics for Governance of Multi-Agent Systems XV",
    pages = "92--108",
    year = "2022",
    publisher = "Springer International Publishing",
    url = "http://dx.doi.org/10.1007/978-3-031-20845-4_6",
    abstract = "In everyday life, people often depend on their
    theory of mind, i.e., their ability to reason about
    unobservable mental content of others to understand,
    explain, and predict their behaviour. Many
    agent-based models have been designed to develop
    computational theory of mind and analyze its
    effectiveness in various tasks and
    settings. However, most existing models are not
    generic (e.g., only applied in a given setting), not
    feasible (e.g., require too much information to be
    processed), or not human-inspired (e.g., do not
    capture the behavioral heuristics of humans). This
    hinders their applicability in many
    settings. Accordingly, we propose a new
    computational theory of mind, which captures the
    human decision heuristics of reasoning by
    abstracting individual beliefs about others. We
    specifically study computational affinity and show
    how it can be used in tandem with theory of mind
    reasoning when designing agent models for
    human-agent negotiation. We perform two-agent
    simulations to analyze the role of affinity in
    getting to agreements when there is a bound on the
    time to be spent for negotiating. Our results
    suggest that modeling affinity can ease the
    negotiation process by decreasing the number of
    rounds needed for an agreement as well as yield a
    higher benefit for agents with theory of mind
    reasoning."
    }

  • M. Michelini, A. Haret, and D. Grossi, “Group Wisdom at a Price: Jury Theorems with Costly Information,” in Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22, 2022, p. 419–425. doi:10.24963/ijcai.2022/60
    [BibTeX] [Abstract] [Download PDF]

    We study epistemic voting on binary issues where voters are characterized by their competence, i.e., the probability of voting for the correct alternative, and can choose between two actions: voting or abstaining. In our setting voting involves the expenditure of some effort, which is required to achieve the appropriate level of competence, whereas abstention carries no effort. We model this scenario as a game and characterize its equilibria under several variations. Our results show that when agents are aware of everyone’s incentives, then the addition of effort may lead to Nash equilibria where wisdom of the crowds is lost. We further show that if agents’ awareness of each other is constrained by a social network, the topology of the network may actually mitigate this effect.

    @inproceedings{michelini22group,
    author = {Michelini, Matteo and Haret, Adrian and Grossi,
    Davide},
    booktitle = {Proceedings of the Thirty-First International Joint
    Conference on Artificial Intelligence, {IJCAI-22}},
    date-added = {2022-12-14 12:31:41 +0100},
    date-modified ={2022-12-14 12:32:24 +0100},
    doi = {10.24963/ijcai.2022/60},
    editor = {Lud De Raedt},
    month = {7},
    note = {Main Track},
    pages = {419--425},
    publisher = {International Joint Conferences on Artificial
    Intelligence Organization},
    title = {Group Wisdom at a Price: Jury Theorems with Costly
    Information},
    url = {https://doi.org/10.24963/ijcai.2022/60},
    year = {2022},
    bdsk-url-1 = {https://doi.org/10.24963/ijcai.2022/60},
    abstract = { We study epistemic voting on binary issues where
    voters are characterized by their competence, i.e.,
    the probability of voting for the correct
    alternative, and can choose between two actions:
    voting or abstaining. In our setting voting involves
    the expenditure of some effort, which is required to
    achieve the appropriate level of competence, whereas
    abstention carries no effort. We model this scenario
    as a game and characterize its equilibria under
    several variations. Our results show that when
    agents are aware of everyone's incentives, then the
    addition of effort may lead to Nash equilibria where
    wisdom of the crowds is lost. We further show that
    if agents' awareness of each other is constrained by
    a social network, the topology of the network may
    actually mitigate this effect. }
    }

  • M. Los, Z. Christoff, and D. Grossi, “Proportional Budget Allocations: Towards a Systematization,” in Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22, 2022, p. 398–404. doi:10.24963/ijcai.2022/57
    [BibTeX] [Abstract] [Download PDF]

    We contribute to the programme of lifting proportionality axioms from the multi-winner voting setting to participatory budgeting. We define novel proportionality axioms for participatory budgeting and test them on known proportionality-driven rules such as Phragmén and Rule X. We investigate logical implications among old and new axioms and provide a systematic overview of proportionality criteria in participatory budgeting.

    @inproceedings{los22proportional,
    author = {Los, Maaike and Christoff, Zoé and Grossi, Davide},
    booktitle = {Proceedings of the Thirty-First International Joint
    Conference on Artificial Intelligence, {IJCAI-22}},
    date-added = {2022-12-14 12:30:46 +0100},
    date-modified ={2022-12-14 12:32:58 +0100},
    doi = {10.24963/ijcai.2022/57},
    editor = {Lud De Raedt},
    month = {7},
    note = {Main Track},
    pages = {398--404},
    publisher = {International Joint Conferences on Artificial
    Intelligence Organization},
    title = {Proportional Budget Allocations: Towards a
    Systematization},
    url = {https://doi.org/10.24963/ijcai.2022/57},
    year = {2022},
    bdsk-url-1 = {https://doi.org/10.24963/ijcai.2022/57},
    abstract = {We contribute to the programme of lifting
    proportionality axioms from the multi-winner voting
    setting to participatory budgeting. We define novel
    proportionality axioms for participatory budgeting
    and test them on known proportionality-driven rules
    such as Phragm{\'e}n and Rule X. We investigate
    logical implications among old and new axioms and
    provide a systematic overview of proportionality
    criteria in participatory budgeting. }
    }

  • Y. Zhang and D. Grossi, “Tracking Truth by Weighting Proxies in Liquid Democracy,” in 21st International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2022, Auckland, New Zealand, May 9-13, 2022, 2022, p. 1482–1490. doi:10.5555/3535850.3536015
    [BibTeX] [Abstract] [Download PDF]

    We study wisdom-of-the-crowd effects in liquid democracy on net- works where agents are allowed to apportion parts of their voting weight to different proxies. We show that in this setting–-unlike in the standard one where voting weight is delegated in full to only one proxy–-it becomes possible to construct delegation struc- tures that optimize the truth-tracking ability of the group. Focusing on group accuracy we contrast this centralized solution with the setting in which agents are free to choose their weighted delega- tions by greedily trying to maximize their own individual accuracy. While equilibria with weighted delegations may be as bad as with standard delegations, they are never worse and may sometimes be better. To gain further insights into this model we experimentally study quantal response delegation strategies on random networks. We observe that weighted delegations can lead, under specific con- ditions, to higher group accuracy than simple majority voting

    @inproceedings{zhang22tracking,
    author = {Yuzhe Zhang and Davide Grossi},
    bibsource = {dblp computer science bibliography,
    https://dblp.org},
    biburl = {https://dblp.org/rec/conf/atal/ZhangG22.bib},
    booktitle = {21st International Conference on Autonomous Agents
    and Multiagent Systems, {AAMAS} 2022, Auckland, New
    Zealand, May 9-13, 2022},
    date-added = {2022-12-14 12:29:16 +0100},
    date-modified ={2022-12-14 12:33:14 +0100},
    doi = {10.5555/3535850.3536015},
    editor = {Piotr Faliszewski and Viviana Mascardi and Catherine
    Pelachaud and Matthew E. Taylor},
    pages = {1482--1490},
    publisher = {International Foundation for Autonomous Agents and
    Multiagent Systems {(IFAAMAS)}},
    timestamp = {Mon, 18 Jul 2022 17:13:00 +0200},
    title = {Tracking Truth by Weighting Proxies in Liquid
    Democracy},
    url =
    {https://www.ifaamas.org/Proceedings/aamas2022/pdfs/p1482.pdf},
    year = {2022},
    abstract = {We study wisdom-of-the-crowd effects in liquid
    democracy on net- works where agents are allowed to
    apportion parts of their voting weight to different
    proxies. We show that in this setting---unlike in
    the standard one where voting weight is delegated in
    full to only one proxy---it becomes possible to
    construct delegation struc- tures that optimize the
    truth-tracking ability of the group. Focusing on
    group accuracy we contrast this centralized solution
    with the setting in which agents are free to choose
    their weighted delega- tions by greedily trying to
    maximize their own individual accuracy. While
    equilibria with weighted delegations may be as bad
    as with standard delegations, they are never worse
    and may sometimes be better. To gain further
    insights into this model we experimentally study
    quantal response delegation strategies on random
    networks. We observe that weighted delegations can
    lead, under specific con- ditions, to higher group
    accuracy than simple majority voting}
    }

  • D. Grossi, W. van der Hoek, and L. B. Kuijer, “Reasoning about General Preference Relations,” Artif. Intell., vol. 313, iss. C, 2022. doi:10.1016/j.artint.2022.103793
    [BibTeX] [Abstract] [Download PDF]

    Preference relations are at the heart of many fundamental concepts in artificial intelligence, ranging from utility comparisons, to defeat among strategies and relative plausibility among states, just to mention a few. Reasoning about such relations has been the object of extensive research and a wealth of formalisms exist to express and reason about them. One such formalism is conditional logic, which focuses on reasoning about the “best” alternatives according to a given preference relation. A “best” alternative is normally interpreted as an alternative that is either maximal (no other alternative is preferred to it) or optimal (it is at least as preferred as all other alternatives). And the preference relation is normally assumed to satisfy strong requirements (typically transitivity and some kind of well-foundedness assumption). Here, we generalize this existing literature in two ways. Firstly, in addition to maximality and optimality, we consider two other interpretations of “best”, which we call unmatchedness and acceptability. Secondly, we do not inherently require the preference relation to satisfy any constraints. Instead, we allow the relation to satisfy any combination of transitivity, totality and anti-symmetry. This allows us to model a wide range of situations, including cases where the lack of constraints stems from a modeled agent being irrational (for example, an agent might have preferences that are neither transitive nor total nor anti-symmetric) or from the interaction of perfectly rational agents (for example, a defeat relation among strategies in a game might be anti-symmetric but not total or transitive). For each interpretation of “best” (maximal, optimal, unmatched or acceptable) and each combination of constraints (transitivity, totality and/or anti-symmetry), we study the sets of valid inferences. Specifically, in all but one case we introduce a sound and strongly complete axiomatization, and in the one remaining case we show that no such axiomatization exists.

    @article{grossi22reasoning,
    address = {GBR},
    author = {Grossi, Davide and van der Hoek, Wiebe and Kuijer,
    Louwe B.},
    date-added = {2022-12-14 12:27:55 +0100},
    date-modified ={2022-12-14 12:33:06 +0100},
    doi = {10.1016/j.artint.2022.103793},
    issn = {0004-3702},
    issue_date = {Dec 2022},
    journal = {Artif. Intell.},
    keywords = {Preference relations, Conditional logic},
    month = {nov},
    number = {C},
    numpages = {44},
    publisher = {Elsevier Science Publishers Ltd.},
    title = {Reasoning about General Preference Relations},
    url = {https://doi.org/10.1016/j.artint.2022.103793},
    volume = {313},
    year = {2022},
    bdsk-url-1 = {https://doi.org/10.1016/j.artint.2022.103793},
    abstract = {Preference relations are at the heart of many
    fundamental concepts in artificial intelligence,
    ranging from utility comparisons, to defeat among
    strategies and relative plausibility among states,
    just to mention a few. Reasoning about such
    relations has been the object of extensive research
    and a wealth of formalisms exist to express and
    reason about them. One such formalism is conditional
    logic, which focuses on reasoning about the ``best''
    alternatives according to a given preference
    relation. A ``best'' alternative is normally
    interpreted as an alternative that is either maximal
    (no other alternative is preferred to it) or optimal
    (it is at least as preferred as all other
    alternatives). And the preference relation is
    normally assumed to satisfy strong requirements
    (typically transitivity and some kind of
    well-foundedness assumption). Here, we generalize
    this existing literature in two ways. Firstly, in
    addition to maximality and optimality, we consider
    two other interpretations of ``best'', which we call
    unmatchedness and acceptability. Secondly, we do not
    inherently require the preference relation to
    satisfy any constraints. Instead, we allow the
    relation to satisfy any combination of transitivity,
    totality and anti-symmetry. This allows us to model
    a wide range of situations, including cases where
    the lack of constraints stems from a modeled agent
    being irrational (for example, an agent might have
    preferences that are neither transitive nor total
    nor anti-symmetric) or from the interaction of
    perfectly rational agents (for example, a defeat
    relation among strategies in a game might be
    anti-symmetric but not total or transitive). For
    each interpretation of ``best'' (maximal, optimal,
    unmatched or acceptable) and each combination of
    constraints (transitivity, totality and/or
    anti-symmetry), we study the sets of valid
    inferences. Specifically, in all but one case we
    introduce a sound and strongly complete
    axiomatization, and in the one remaining case we
    show that no such axiomatization exists.}
    }

  • A. Onnes, Monitoring AI Systems: A Problem Analysis, Framework and OutlookIOS press, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Knowledge-based systems have been used to monitor machines and processes in the real world. In this paper we propose the use of knowledge-based systems to monitor other AI systems in operation. We motivate and provide a problem analysis of this novel setting and subsequently propose a framework that allows for structuring future research related to this setting. Several directions for further research are also discussed.

    @proceedings{Onnes2022,
    author = {Onnes, Annet},
    title = {Monitoring AI Systems: A Problem Analysis, Framework
    and Outlook},
    year = {2022},
    booktitle = {HHAI 2022: Augmenting Human Intellect},
    editor = {Schlobach, Stefan and Pérez-Ortiz, María and
    Tielman, Myrthe},
    volume = {354},
    series = {Frontiers in Artificial Intelligence and
    Applications},
    publisher = {IOS press},
    URL = {https://ebooks.iospress.nl/volumearticle/60870},
    Abstract = {Knowledge-based systems have been used to monitor
    machines and processes in the real world. In this
    paper we propose the use of knowledge-based systems
    to monitor other AI systems in operation. We
    motivate and provide a problem analysis of this
    novel setting and subsequently propose a framework
    that allows for structuring future research related
    to this setting. Several directions for further
    research are also discussed.}
    }

  • K. Deja, A. Kuzina, T. Trzciński, and J. M. Tomczak, “On Analyzing Generative and Denoising Capabilities of Diffusion-based Deep Generative Models,” 36th Conference on Neural Information Processing Systems (NeurIPS 2022), 2022.
    [BibTeX] [Abstract] [Download PDF]

    Diffusion-based Deep Generative Models (DDGMs) offer state-of-the-art performance in generative modeling. Their main strength comes from their unique setup in which a model (the backward diffusion process) is trained to reverse the forward diffusion process, which gradually adds noise to the input signal. Although DDGMs are well studied, it is still unclear how the small amount of noise is transformed during the backward diffusion process. Here, we focus on analyzing this problem to gain more insight into the behavior of DDGMs and their denoising and generative capabilities. We observe a fluid transition point that changes the functionality of the backward diffusion process from generating a (corrupted) image from noise to denoising the corrupted image to the final sample. Based on this observation, we postulate to divide a DDGM into two parts: a denoiser and a generator. The denoiser could be parameterized by a denoising auto-encoder, while the generator is a diffusion-based model with its own set of parameters. We experimentally validate our proposition, showing its pros and cons.

    @article{deja2022analyzing,
    title = {On Analyzing Generative and Denoising Capabilities
    of Diffusion-based Deep Generative Models},
    author = {Deja, Kamil and Kuzina, Anna and Trzci{\'n}ski,
    Tomasz and Tomczak, Jakub M},
    journal = {36th Conference on Neural Information Processing
    Systems (NeurIPS 2022)},
    year = {2022},
    url = {https://arxiv.org/abs/2206.00070},
    abstract = {Diffusion-based Deep Generative Models (DDGMs) offer
    state-of-the-art performance in generative
    modeling. Their main strength comes from their
    unique setup in which a model (the backward
    diffusion process) is trained to reverse the forward
    diffusion process, which gradually adds noise to the
    input signal. Although DDGMs are well studied, it is
    still unclear how the small amount of noise is
    transformed during the backward diffusion
    process. Here, we focus on analyzing this problem to
    gain more insight into the behavior of DDGMs and
    their denoising and generative capabilities. We
    observe a fluid transition point that changes the
    functionality of the backward diffusion process from
    generating a (corrupted) image from noise to
    denoising the corrupted image to the final
    sample. Based on this observation, we postulate to
    divide a DDGM into two parts: a denoiser and a
    generator. The denoiser could be parameterized by a
    denoising auto-encoder, while the generator is a
    diffusion-based model with its own set of
    parameters. We experimentally validate our
    proposition, showing its pros and cons.}
    }

  • A. Kuzina, M. Welling, and J. M. Tomczak, “Alleviating Adversarial Attacks on Variational Autoencoders with MCMC,” 36th Conference on Neural Information Processing Systems (NeurIPS 2022), 2022.
    [BibTeX] [Abstract] [Download PDF]

    Variational autoencoders (VAEs) are latent variable models that can generate complex objects and provide meaningful latent representations. Moreover, they could be further used in downstream tasks such as classification. As previous work has shown, one can easily fool VAEs to produce unexpected latent representations and reconstructions for a visually slightly modified input. Here, we examine several objective functions for adversarial attack construction proposed previously and present a solution to alleviate the effect of these attacks. Our method utilizes the Markov Chain Monte Carlo (MCMC) technique in the inference step that we motivate with a theoretical analysis. Thus, we do not incorporate any extra costs during training, and the performance on non-attacked inputs is not decreased. We validate our approach on a variety of datasets (MNIST, Fashion MNIST, Color MNIST, CelebA) and VAE configurations (β -VAE, NVAE, β-TCVAE), and show that our approach consistently improves the model robustness to adversarial attacks.

    @article{kuzina2022alleviating,
    title = {Alleviating Adversarial Attacks on Variational
    Autoencoders with MCMC},
    author = {Kuzina, Anna and Welling, Max and Tomczak, Jakub M},
    journal = {36th Conference on Neural Information Processing
    Systems (NeurIPS 2022)},
    year = {2022},
    url = {https://arxiv.org/abs/2203.09940},
    abstract = {Variational autoencoders (VAEs) are latent variable
    models that can generate complex objects and provide
    meaningful latent representations. Moreover, they
    could be further used in downstream tasks such as
    classification. As previous work has shown, one can
    easily fool VAEs to produce unexpected latent
    representations and reconstructions for a visually
    slightly modified input. Here, we examine several
    objective functions for adversarial attack
    construction proposed previously and present a
    solution to alleviate the effect of these
    attacks. Our method utilizes the Markov Chain Monte
    Carlo (MCMC) technique in the inference step that we
    motivate with a theoretical analysis. Thus, we do
    not incorporate any extra costs during training, and
    the performance on non-attacked inputs is not
    decreased. We validate our approach on a variety of
    datasets (MNIST, Fashion MNIST, Color MNIST, CelebA)
    and VAE configurations (β -VAE, NVAE, β-TCVAE), and
    show that our approach consistently improves the
    model robustness to adversarial attacks.}
    }

  • S. Vadgama, J. M. Tomczak, and E. Bekkers, “Kendall Shape-VAE : Learning Shapes in a Generative Framework,” in NeurIPS 2022 Workshop on Symmetry and Geometry in Neural Representations, 2022.
    [BibTeX] [Abstract] [Download PDF]

    Learning an interpretable representation of data without supervision is an important precursor for the development of artificial intelligence. In this work, we introduce \textit{Kendall Shape}-VAE, a novel Variational Autoencoder framework for learning shapes as it disentangles the latent space by compressing information to simpler geometric symbols. In \textit{Kendall Shape}-VAE, we modify the Hyperspherical Variational Autoencoder such that it results in an exactly rotationally equivariant network using the notion of landmarks in the Kendall shape space. We show the exact equivariance of the model through experiments on rotated MNIST.

    @inproceedings{vadgama2022kendall,
    title = {Kendall Shape-{VAE} : Learning Shapes in a
    Generative Framework},
    author = {Sharvaree Vadgama and Jakub Mikolaj Tomczak and Erik
    J Bekkers},
    booktitle = {NeurIPS 2022 Workshop on Symmetry and Geometry in
    Neural Representations},
    year = {2022},
    url = {https://openreview.net/forum?id=nzh4N6kdl2G},
    abstract = {Learning an interpretable representation of data
    without supervision is an important precursor for
    the development of artificial intelligence. In this
    work, we introduce \textit{Kendall Shape}-VAE, a
    novel Variational Autoencoder framework for learning
    shapes as it disentangles the latent space by
    compressing information to simpler geometric
    symbols. In \textit{Kendall Shape}-VAE, we modify
    the Hyperspherical Variational Autoencoder such that
    it results in an exactly rotationally equivariant
    network using the notion of landmarks in the Kendall
    shape space. We show the exact equivariance of the
    model through experiments on rotated MNIST.}
    }

  • E. Liscio, A. E. Dondera, A. Geadau, C. M. Jonker, and P. K. Murukannaiah, “Cross-Domain Classification of Moral Values,” in Findings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics, Seattle, WA, USA, 2022, p. 2727–2745.
    [BibTeX] [Abstract] [Download PDF]

    Moral values influence how we interpret and act upon the information we receive. Identifying human moral values is essential for artificially intelligent agents to co-exist with humans. Recent progress in natural language processing allows the identification of moral values in textual discourse. However, domain-specific moral rhetoric poses challenges for transferring knowledge from one domain to another.We provide the first extensive investigation on the effects of cross-domain classification of moral values from text. We compare a state-of-the-art deep learning model (BERT) in seven domains and four cross-domain settings. We show that a value classifier can generalize and transfer knowledge to novel domains, but it can introduce catastrophic forgetting. We also highlight the typical classification errors in cross-domain value classification and compare the model predictions to the annotators agreement. Our results provide insights to computer and social scientists that seek to identify moral rhetoric specific to a domain of discourse.

    @inproceedings{Liscio2022a,
    title = {{Cross-Domain Classification of Moral Values}},
    year = {2022},
    booktitle = {Findings of the 2022 Conference of the North
    American Chapter of the Association for
    Computational Linguistics},
    series = {NAACL '22},
    author = {Liscio, Enrico and Dondera, Alin E. and Geadau,
    Andrei and Jonker, Catholijn M. and Murukannaiah,
    Pradeep K.},
    pages = {2727--2745},
    publisher = {ACL},
    address = {Seattle, WA, USA},
    url =
    {https://aclanthology.org/2022.findings-naacl.209.pdf},
    abstract = {Moral values influence how we interpret and act upon
    the information we receive. Identifying human moral
    values is essential for artificially intelligent
    agents to co-exist with humans. Recent progress in
    natural language processing allows the
    identification of moral values in textual
    discourse. However, domain-specific moral rhetoric
    poses challenges for transferring knowledge from one
    domain to another.We provide the first extensive
    investigation on the effects of cross-domain
    classification of moral values from text. We compare
    a state-of-the-art deep learning model (BERT) in
    seven domains and four cross-domain settings. We
    show that a value classifier can generalize and
    transfer knowledge to novel domains, but it can
    introduce catastrophic forgetting. We also highlight
    the typical classification errors in cross-domain
    value classification and compare the model
    predictions to the annotators agreement. Our results
    provide insights to computer and social scientists
    that seek to identify moral rhetoric specific to a
    domain of discourse.}
    }

  • L. C. Siebert, E. Liscio, P. K. Murukannaiah, L. Kaptein, S. L. Spruit, J. van den Hoven, and C. M. Jonker, “Estimating Value Preferences in a Hybrid Participatory System,” in HHAI2022: Augmenting Human Intellect, Amsterdam, the Netherlands, 2022, p. 114–127.
    [BibTeX] [Abstract] [Download PDF]

    We propose methods for an AI agent to estimate the value preferences of individuals in a hybrid participatory system, considering a setting where participants make choices and provide textual motivations for those choices. We focus on situations where there is a conflict between participants’ choices and motivations, and operationalize the philosophical stance that “valuing is deliberatively consequential. That is, if a user’s choice is based on a deliberation of value preferences, the value preferences can be observed in the motivation the user provides for the choice. Thus, we prioritize the value preferences estimated from motivations over the value preferences estimated from choices alone. We evaluate the proposed methods on a dataset of a large-scale survey on energy transition. The results show that explicitly addressing inconsistencies between choices and motivations improves the estimation of an individual’s value preferences. The proposed methods can be integrated in a hybrid participatory system, where artificial agents ought to estimate humans’ value preferences to pursue value alignment.

    @inproceedings{Siebert2022,
    title = {{Estimating Value Preferences in a Hybrid
    Participatory System}},
    year = {2022},
    booktitle = {HHAI2022: Augmenting Human Intellect},
    author = {Siebert, Luciano C. and Liscio, Enrico and
    Murukannaiah, Pradeep K. and Kaptein, Lionel and
    Spruit, Shannon L. and van den Hoven, Jeroen and
    Jonker, Catholijn M.},
    pages = {114--127},
    publisher = {IOS Press},
    series = {HHAI '22},
    address = {Amsterdam, the Netherlands},
    url = {https://ebooks.iospress.nl/volumearticle/60861},
    abstract = {We propose methods for an AI agent to estimate the
    value preferences of individuals in a hybrid
    participatory system, considering a setting where
    participants make choices and provide textual
    motivations for those choices. We focus on
    situations where there is a conflict between
    participants' choices and motivations, and
    operationalize the philosophical stance that
    “valuing is deliberatively consequential. That is,
    if a user's choice is based on a deliberation of
    value preferences, the value preferences can be
    observed in the motivation the user provides for the
    choice. Thus, we prioritize the value preferences
    estimated from motivations over the value
    preferences estimated from choices alone. We
    evaluate the proposed methods on a dataset of a
    large-scale survey on energy transition. The results
    show that explicitly addressing inconsistencies
    between choices and motivations improves the
    estimation of an individual's value preferences. The
    proposed methods can be integrated in a hybrid
    participatory system, where artificial agents ought
    to estimate humans' value preferences to pursue
    value alignment.}
    }

  • M. van der Meer, E. Liscio, C. M. Jonker, A. Plaat, P. Vossen, and P. K. Murukannaiah, “HyEnA: A Hybrid Method for Extracting Arguments from Opinions,” in HHAI2022: Augmenting Human Intellect, Amsterdam, the Netherlands, 2022, p. 17–31.
    [BibTeX] [Abstract] [Download PDF]

    The key arguments underlying a large and noisy set of opinions help understand the opinions quickly and accurately. Fully automated methods can extract arguments but (1) require large labeled datasets and (2) work well for known viewpoints, but not for novel points of view. We propose HyEnA, a hybrid (human + AI) method for extracting arguments from opinionated texts, combining the speed of automated processing with the understanding and reasoning capabilities of humans. We evaluate HyEnA on three feedback corpora. We find that, on the one hand, HyEnA achieves higher coverage and precision than a state-of-the-art automated method, when compared on a common set of diverse opinions, justifying the need for human insight. On the other hand, HyEnA requires less human effort and does not compromise quality compared to (fully manual) expert analysis, demonstrating the benefit of combining human and machine intelligence.

    @inproceedings{vanderMeer2022,
    title = {{HyEnA: A Hybrid Method for Extracting Arguments
    from Opinions}},
    year = {2022},
    booktitle = {HHAI2022: Augmenting Human Intellect},
    author = {van der Meer, Michiel and Liscio, Enrico and Jonker,
    Catholijn M. and Plaat, Aske and Vossen, Piek and
    Murukannaiah, Pradeep K.},
    pages = {17--31},
    publisher = {IOS Press},
    series = {HHAI '22},
    address = {Amsterdam, the Netherlands},
    url = {https://ebooks.iospress.nl/volumearticle/60855},
    abstract = {The key arguments underlying a large and noisy set
    of opinions help understand the opinions quickly and
    accurately. Fully automated methods can extract
    arguments but (1) require large labeled datasets and
    (2) work well for known viewpoints, but not for
    novel points of view. We propose HyEnA, a hybrid
    (human + AI) method for extracting arguments from
    opinionated texts, combining the speed of automated
    processing with the understanding and reasoning
    capabilities of humans. We evaluate HyEnA on three
    feedback corpora. We find that, on the one hand,
    HyEnA achieves higher coverage and precision than a
    state-of-the-art automated method, when compared on
    a common set of diverse opinions, justifying the
    need for human insight. On the other hand, HyEnA
    requires less human effort and does not compromise
    quality compared to (fully manual) expert analysis,
    demonstrating the benefit of combining human and
    machine intelligence.}
    }

  • R. Loftin and F. A. Oliehoek, “On the Impossibility of Learning to Cooperate with Adaptive Partner Strategies in Repeated Games,” in International Conference on Machine Learning, 2022, p. 14197–14209.
    [BibTeX] [Abstract] [Download PDF]

    Learning to cooperate with other agents is challenging when those agents also possess the ability to adapt to our own behavior. Practical and theoretical approaches to learning in cooperative settings typically assume that other agents’ behaviors are stationary, or else make very specific assumptions about other agents’ learning processes. The goal of this work is to understand whether we can reliably learn to cooperate with other agents without such restrictive assumptions, which are unlikely to hold in real-world applications. Our main contribution is a set of impossibility results, which show that no learning algorithm can reliably learn to cooperate with all possible adaptive partners in a repeated matrix game, even if that partner is guaranteed to cooperate with some stationary strategy. Motivated by these results, we then discuss potential alternative assumptions which capture the idea that an adaptive partner will only adapt rationally to our behavior.

    @inproceedings{loftin2022impossibility,
    title = {On the Impossibility of Learning to Cooperate with
    Adaptive Partner Strategies in Repeated Games},
    author = {Loftin, Robert and Oliehoek, Frans A},
    booktitle = {International Conference on Machine Learning},
    pages = {14197--14209},
    year = {2022},
    organization = {PMLR},
    abstract = {Learning to cooperate with other agents is
    challenging when those agents also possess the
    ability to adapt to our own behavior. Practical and
    theoretical approaches to learning in cooperative
    settings typically assume that other agents'
    behaviors are stationary, or else make very specific
    assumptions about other agents' learning
    processes. The goal of this work is to understand
    whether we can reliably learn to cooperate with
    other agents without such restrictive assumptions,
    which are unlikely to hold in real-world
    applications. Our main contribution is a set of
    impossibility results, which show that no learning
    algorithm can reliably learn to cooperate with all
    possible adaptive partners in a repeated matrix
    game, even if that partner is guaranteed to
    cooperate with some stationary strategy. Motivated
    by these results, we then discuss potential
    alternative assumptions which capture the idea that
    an adaptive partner will only adapt rationally to
    our behavior.},
    URL = {https://arxiv.org/abs/2206.10614}
    }

  • J. van Rhenen, C. Centeio Jorge, T. Matej Hrkalovic, and B. Dudzik, “Effects of Social Behaviours in Online Video Games on Team Trust,” in Extended Abstracts of the 2022 Annual Symposium on Computer-Human Interaction in Play, 2022, p. 159–165.
    [BibTeX] [Abstract] [Download PDF]

    In competitive multiplayer online video games, teamwork is of utmost importance, implying high levels of interdependence between the joint outcomes of players. When engaging in such interdependent interactions, humans rely on trust to facilitate coordination of their individual behaviours. However, online games often take place between teams of strangers, with individual members having little to no information about each other than what they observe throughout the interaction itself. A better understanding of the social behaviours that are used by players to form trust could not only facilitate richer gaming experiences, but could also lead to insights about team interactions. As such, this paper presents a first step towards understanding how and which types of in-game behaviour relate to trust formation. In particular, we investigate a)which in-game behaviour were relevant for trust formation (first part of the study) and b) how they relate to the reported player’s trust in their teammates (the second part of the study). The first part consisted of interviews with League of Legends players in order to create a taxonomy of in-game behaviours relevant for trust formation. As for the second part, we ran a small-scale pilot study where participants played the game and then answered a questionnaire to measure their trust in their teammates. Our preliminary results present a taxonomy of in-game behaviours which can be used to annotate the games regarding trust behaviours. Based on the pilot study, the list of behaviours could be extended as to improve the results. These findings can be used to research the role of trust formation in teamwork

    @inproceedings{van2022effects,
    title={Effects of Social Behaviours in Online Video Games on Team Trust},
    author={van Rhenen, Jan-Willem and Centeio Jorge, Carolina and Matej Hrkalovic, Tiffany and Dudzik, Bernd},
    booktitle={Extended Abstracts of the 2022 Annual Symposium on Computer-Human Interaction in Play},
    pages={159--165},
    year={2022},
    URL={https://pure.tudelft.nl/admin/files/146989190/vanRhenen2021_author.pdf},
    Abstract={In competitive multiplayer online video games, teamwork is of utmost importance, implying high levels of interdependence between the joint outcomes of players. When engaging in such interdependent interactions, humans rely on trust to facilitate coordination of their individual behaviours. However, online games often take place between teams of strangers, with individual members having little to no information about each other than what they observe throughout the interaction itself. A better understanding of the social behaviours that are used by players to form trust could not only facilitate richer gaming experiences, but could also lead to insights about team interactions. As such, this paper presents a first step towards understanding how and which types of in-game behaviour relate to trust formation. In particular, we investigate a)which in-game behaviour were relevant for trust formation (first part of the study) and b) how they relate to the reported player’s trust in their teammates (the second part of the study). The first part consisted of interviews with League of Legends players in order to create a taxonomy of in-game behaviours relevant for trust formation. As for the second part, we ran a small-scale pilot study where participants played the game and then answered a questionnaire to measure their trust in their teammates. Our preliminary results present a taxonomy of in-game behaviours which can be used to annotate the games regarding trust behaviours. Based on the pilot study, the list of behaviours could be extended as to improve the results. These findings can be used to research the role of trust formation in teamwork}
    }

  • T. Matej Hrkalovic, “Designing Hybrid Intelligence Techniques for Facilitating Collaboration Informed by Social Science,” in Proceedings of the 2022 International Conference on Multimodal Interaction, 2022, p. 679–684.
    [BibTeX] [Abstract] [Download PDF]

    Designing (socially) intelligent systems for facilitating collaborations in human-human and human-AI teams will require them to have a basic understanding of principles underlying social decision- making. Partner selection – the ability to identify and select suitable partners for collaborative relationships – is one relevant component of social intelligence and an important ingredient for successful relationship management. In everyday life, decision to engage in joint undertakings are often based on impressions made during social interactions with potential partners. These impressions, and consequently, partner selection are informed by (non)-verbal behavioral cues. Despite its importance, research investigating how these impressions and partner selection decisions unfold in naturalistic settings seem to be lacking. Thus, in this paper, we present a project focused on understanding, predicting and modeling partner selection and understanding its relationship with human impressions in semi- naturalistic settings, such as social interactions, with the aim of informing future designing approaches of (hybrid) intelligence system that can understand, predict and aid in initiating and facilitating (current and future) collaborations.

    @inproceedings{matej2022designing,
    title={Designing Hybrid Intelligence Techniques for Facilitating Collaboration Informed by Social Science},
    author={Matej Hrkalovic, Tiffany},
    booktitle={Proceedings of the 2022 International Conference on Multimodal Interaction},
    pages={679--684},
    year={2022},
    URL={https://research.vu.nl/en/publications/designing-hybrid-intelligence-techniques-for-facilitating-collabo},
    Abstract={Designing (socially) intelligent systems for facilitating collaborations in human-human and human-AI teams will require them to have a basic understanding of principles underlying social decision-
    making. Partner selection - the ability to identify and select suitable partners for collaborative relationships - is one relevant component of social intelligence and an important ingredient for successful relationship management. In everyday life, decision to engage in
    joint undertakings are often based on impressions made during social interactions with potential partners. These impressions, and consequently, partner selection are informed by (non)-verbal behavioral cues. Despite its importance, research investigating how these impressions and partner selection decisions unfold in naturalistic settings seem to be lacking. Thus, in this paper, we present a project focused on understanding, predicting and modeling partner selection and understanding its relationship with human impressions in semi- naturalistic settings, such as social interactions, with the aim of informing future designing approaches of (hybrid) intelligence system that can understand, predict and aid in initiating and facilitating (current and future) collaborations.}
    }

  • U. Khurana, I. Vermeulen, E. Nalisnick, M. Van Noorloos, and A. Fokkens, “Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions,” in Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH), Seattle, Washington (Hybrid), 2022, p. 176–191. doi:10.18653/v1/2022.woah-1.17
    [BibTeX] [Abstract] [Download PDF]

    The subjectivity of automatic hate speech detection makes it a complex task, reflected in different and incomplete definitions in NLP. We present hate speech criteria, developed with insights from a law and social science expert, that help researchers create more explicit definitions and annotation guidelines on five aspects: (1) target groups and (2) dominance, (3) perpetrator characteristics, (4) explicit presence of negative interactions, and the (5) type of consequences/effects. Definitions can be structured so that they cover a more broad or more narrow phenomenon and conscious choices can be made on specifying criteria or leaving them open. We argue that the goal and exact task developers have in mind should determine how the scope of hate speech is defined. We provide an overview of the properties of datasets from hatespeechdata.com that may help select the most suitable dataset for a specific scenario.

    @inproceedings{khurana-etal-2022-hate,
    title = "Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions",
    author = "Khurana, Urja and
    Vermeulen, Ivar and
    Nalisnick, Eric and
    Van Noorloos, Marloes and
    Fokkens, Antske",
    booktitle = "Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH)",
    month = "jul",
    year = "2022",
    address = "Seattle, Washington (Hybrid)",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2022.woah-1.17",
    doi = "10.18653/v1/2022.woah-1.17",
    pages = "176--191",
    abstract = "The subjectivity of automatic hate speech detection makes it a complex task, reflected in different and incomplete definitions in NLP. We present hate speech criteria, developed with insights from a law and social science expert, that help researchers create more explicit definitions and annotation guidelines on five aspects: (1) target groups and (2) dominance, (3) perpetrator characteristics, (4) explicit presence of negative interactions, and the (5) type of consequences/effects. Definitions can be structured so that they cover a more broad or more narrow phenomenon and conscious choices can be made on specifying criteria or leaving them open. We argue that the goal and exact task developers have in mind should determine how the scope of hate speech is defined. We provide an overview of the properties of datasets from hatespeechdata.com that may help select the most suitable dataset for a specific scenario.",
    }

2021

  • M. van Bekkum, M. de Boer, F. van Harmelen, A. M. -, and A. Teije, “Modular design patterns for hybrid learning and reasoning systems,” Appl. Intell., vol. 51, iss. 9, p. 6528–6546, 2021. doi:10.1007/s10489-021-02394-3
    [BibTeX] [Abstract] [Download PDF]

    The unification of statistical (data-driven) and symbolic (knowledge-driven) methods is widely recognised as one of the key challenges of modern AI. Recent years have seen large number of publications on such hybrid neuro-symbolic AI systems. That rapidly growing literature is highly diverse and mostly empirical, and is lacking a unifying view of the large variety of these hybrid systems. In this paper we analyse a large body of recent literature and we propose a set of modular design patterns for such hybrid, neuro-symbolic systems. We are able to describe the architecture of a very large number of hybrid systems by composing only a small set of elementary patterns as building blocks. The main contributions of this paper are: 1) a taxonomically organised vocabulary to describe both processes and data structures used in hybrid systems; 2) a set of 15+ design patterns for hybrid AI systems, organised in a set of elementary patterns and a set of compositional patterns; 3) an application of these design patterns in two realistic use-cases for hybrid AI systems. Our patterns reveal similarities between systems that were not recognised until now. Finally, our design patterns extend and refine Kautz’ earlier attempt at categorising neuro-symbolic architectures.

    @article{DBLP:journals/apin/BekkumBHMT21,
    author = {Michael van Bekkum and Maaike de Boer and Frank van
    Harmelen and Andr{\'{e}} Meyer{-}Vitali and Annette
    ten Teije},
    title = {Modular design patterns for hybrid learning and
    reasoning systems},
    journal = {Appl. Intell.},
    volume = {51},
    number = {9},
    pages = {6528--6546},
    year = {2021},
    url = {https://doi.org/10.1007/s10489-021-02394-3},
    doi = {10.1007/s10489-021-02394-3},
    timestamp = {Wed, 01 Sep 2021 12:45:13 +0200},
    biburl = {https://dblp.org/rec/journals/apin/BekkumBHMT21.bib},
    url =
    {https://link.springer.com/article/10.1007/s10489-021-02394-3
    },
    abstract = {The unification of statistical (data-driven) and
    symbolic (knowledge-driven) methods is widely
    recognised as one of the key challenges of modern
    AI. Recent years have seen large number of
    publications on such hybrid neuro-symbolic AI
    systems. That rapidly growing literature is highly
    diverse and mostly empirical, and is lacking a
    unifying view of the large variety of these hybrid
    systems. In this paper we analyse a large body of
    recent literature and we propose a set of modular
    design patterns for such hybrid, neuro-symbolic
    systems. We are able to describe the architecture of
    a very large number of hybrid systems by composing
    only a small set of elementary patterns as building
    blocks. The main contributions of this paper are:
    1) a taxonomically organised vocabulary to describe
    both processes and data structures used in hybrid
    systems; 2) a set of 15+ design patterns for hybrid
    AI systems, organised in a set of elementary
    patterns and a set of compositional patterns; 3) an
    application of these design patterns in two
    realistic use-cases for hybrid AI systems. Our
    patterns reveal similarities between systems that
    were not recognised until now. Finally, our design
    patterns extend and refine Kautz' earlier attempt at
    categorising neuro-symbolic architectures.}
    }

  • A. Kuzina, M. Welling, and J. M. Tomczak, “Diagnosing Vulnerability of Variational Auto-Encoders to Adversarial Attacks,” in ICLR 2021 Workshop on Robust and Reliable Machine Learning in the Real World, 2021.
    [BibTeX] [Abstract] [Download PDF]

    In this work, we explore adversarial attacks on the Variational Autoencoders (VAE). We show how to modify data point to obtain a prescribed latent code (supervised attack) or just get a drastically different code (unsupervised attack). We examine the influence of model modifications ($\beta$-VAE, NVAE) on the robustness of VAEs and suggest metrics to quantify it.

    @inproceedings{kuzina2021diagnosing,
    title = {Diagnosing Vulnerability of Variational
    Auto-Encoders to Adversarial Attacks},
    author = {Kuzina, Anna and Welling, Max and Tomczak, Jakub M},
    year = {2021},
    booktitle = {ICLR 2021 Workshop on Robust and Reliable Machine
    Learning in the Real World},
    url = {https://arxiv.org/pdf/2103.06701.pdf},
    abstract = {In this work, we explore adversarial attacks on the
    Variational Autoencoders (VAE). We show how to
    modify data point to obtain a prescribed latent code
    (supervised attack) or just get a drastically
    different code (unsupervised attack). We examine the
    influence of model modifications ($\beta$-VAE, NVAE)
    on the robustness of VAEs and suggest metrics to
    quantify it.}
    }

  • H. Zheng and B. Verheij, “Rules, cases and arguments in artificial intelligence and law,” in Research Handbook on Big Data Law, R. Vogl, Ed., Edgar Elgar Publishing, 2021, pp. 373-387.
    [BibTeX] [Abstract] [Download PDF]

    Artificial intelligence and law is an interdisciplinary field of research that dates back at least to the 1970s, with academic conferences starting in the 1980s. In the field, complex problems are addressed about the computational modeling and automated support of legal reasoning and argumentation. Scholars have different backgrounds, and progress is driven by insights from lawyers, judges, computer scientists, philosophers and others. The community investigates and develops artificial intelligence techniques applicable in the legal domain, in order to enhance access to law for citizens and to support the efficiency and quality of work in the legal domain, aiming to promote a just society. Integral to the legal domain, legal reasoning and its structure and process have gained much attention in AI & Law research. Such research is today especially relevant, since in these days of big data and widespread use of algorithms, there is a need in AI to connect knowledge-based and data-driven AI techniques in order to arrive at a social, explainable and responsible AI. By considering knowledge in the form of rules and data in the form of cases connected by arguments, the field of AI & Law contributes relevant representations and algorithms for handling a combination of knowledge and data. In this chapter, as an entry point into the literature on AI & Law, three major styles of modeling legal reasoning are studied: rule-based reasoning, case-based reasoning and argument-based reasoning, which are the focus of this chapter. We describe selected key ideas, leaving out formal detail. As we will see, these styles of modeling legal reasoning are related, and there is much research investigating relations. We use the example domain of Dutch tort law (Section 2) to illustrate these three major styles, which are then more fully explained (Sections 3 to 5)

    @InCollection{Zheng:2021,
    author = {H. Zheng and B. Verheij},
    title = {Rules, cases and arguments in artificial
    intelligence and law},
    booktitle = {Research Handbook on Big Data Law},
    publisher = {Edgar Elgar Publishing},
    editor = {R Vogl},
    year = 2021,
    url =
    {https://www.ai.rug.nl/~verheij/publications/handbook2021.htm},
    pages = {373-387},
    abstract = {Artificial intelligence and law is an
    interdisciplinary field of research that dates back
    at least to the 1970s, with academic conferences
    starting in the 1980s. In the field, complex
    problems are addressed about the computational
    modeling and automated support of legal reasoning
    and argumentation. Scholars have different
    backgrounds, and progress is driven by insights from
    lawyers, judges, computer scientists, philosophers
    and others. The community investigates and develops
    artificial intelligence techniques applicable in the
    legal domain, in order to enhance access to law for
    citizens and to support the efficiency and quality
    of work in the legal domain, aiming to promote a
    just society. Integral to the legal domain, legal
    reasoning and its structure and process have gained
    much attention in AI & Law research. Such research
    is today especially relevant, since in these days of
    big data and widespread use of algorithms, there is
    a need in AI to connect knowledge-based and
    data-driven AI techniques in order to arrive at a
    social, explainable and responsible AI. By
    considering knowledge in the form of rules and data
    in the form of cases connected by arguments, the
    field of AI & Law contributes relevant
    representations and algorithms for handling a
    combination of knowledge and data. In this chapter,
    as an entry point into the literature on AI & Law,
    three major styles of modeling legal reasoning are
    studied: rule-based reasoning, case-based reasoning
    and argument-based reasoning, which are the focus of
    this chapter. We describe selected key ideas,
    leaving out formal detail. As we will see, these
    styles of modeling legal reasoning are related, and
    there is much research investigating relations. We
    use the example domain of Dutch tort law (Section 2)
    to illustrate these three major styles, which are
    then more fully explained (Sections 3 to 5)}
    }

  • C. A. Kurtan and P. i, “Assisting humans in privacy management: an agent-based approach,” Autonomous Agents and Multi-Agent Systems, vol. 35, iss. 7, 2021. doi:https://doi.org/10.1007/s10458-020-09488-1
    [BibTeX] [Abstract] [Download PDF]

    Image sharing is a service offered by many online social networks. In order to preserve privacy of images, users need to think through and specify a privacy setting for each image that they upload. This is difficult for two main reasons: first, research shows that many times users do not know their own privacy preferences, but only become aware of them over time. Second, even when users know their privacy preferences, editing these privacy settings is cumbersome and requires too much effort, interfering with the quick sharing behavior expected on an online social network. Accordingly, this paper proposes a privacy recommendation model for images using tags and an agent that implements this, namely pelte. Each user agent makes use of the privacy settings that its user have set for previous images to predict automatically the privacy setting for an image that is uploaded to be shared. When in doubt, the agent analyzes the sharing behavior of other users in the user’s network to be able to recommend to its user about what should be considered as private. Contrary to existing approaches that assume all the images are available to a centralized model, pelte is compatible to distributed environments since each agent accesses only the privacy settings of the images that the agent owner has shared or those that have been shared with the user. Our simulations on a real-life dataset shows that pelte can accurately predict privacy settings even when a user has shared a few images with others, the images have only a few tags or the user’s friends have varying privacy preferences.

    @Article{kurtan-yolum-21,
    author = {A. Can Kurtan and P{\i}nar Yolum},
    title = {Assisting humans in privacy management: an
    agent-based approach},
    journal = {Autonomous Agents and Multi-Agent Systems},
    year = {2021},
    volume = {35},
    number = {7},
    abstract = {Image sharing is a service offered by many online
    social networks. In order to preserve privacy of
    images, users need to think through and specify a
    privacy setting for each image that they
    upload. This is difficult for two main reasons:
    first, research shows that many times users do not
    know their own privacy preferences, but only become
    aware of them over time. Second, even when users
    know their privacy preferences, editing these
    privacy settings is cumbersome and requires too much
    effort, interfering with the quick sharing behavior
    expected on an online social network. Accordingly,
    this paper proposes a privacy recommendation model
    for images using tags and an agent that implements
    this, namely pelte. Each user agent makes use of the
    privacy settings that its user have set for previous
    images to predict automatically the privacy setting
    for an image that is uploaded to be shared. When in
    doubt, the agent analyzes the sharing behavior of
    other users in the user's network to be able to
    recommend to its user about what should be
    considered as private. Contrary to existing
    approaches that assume all the images are available
    to a centralized model, pelte is compatible to
    distributed environments since each agent accesses
    only the privacy settings of the images that the
    agent owner has shared or those that have been
    shared with the user. Our simulations on a real-life
    dataset shows that pelte can accurately predict
    privacy settings even when a user has shared a few
    images with others, the images have only a few tags
    or the user's friends have varying privacy
    preferences.},
    url =
    {https://link.springer.com/article/10.1007/s10458-020-09488-1},
    doi = {https://doi.org/10.1007/s10458-020-09488-1}
    }

  • E. Liscio, M. van der Meer, C. M. Jonker, and P. K. Murukannaiah, “A Collaborative Platform for Identifying Context-Specific Values,” in Proceedings of the 20th International Conference on Autonomous Agents and Multiagent Systems, Online, 2021, p. 1773–1775.
    [BibTeX] [Abstract] [Download PDF]

    Value alignment is a crucial aspect of ethical multiagent systems. An important step toward value alignment is identifying values specific to an application context. However, identifying context-specific values is complex and cognitively demanding. To support this process, we develop a methodology and a collaborative web platform that employs AI techniques. We describe this platform, highlighting its intuitive design and implementation.

    @inproceedings{Liscio2021a,
    title = {{A Collaborative Platform for Identifying
    Context-Specific Values}},
    year = {2021},
    booktitle = {Proceedings of the 20th International Conference on
    Autonomous Agents and Multiagent Systems},
    series = {AAMAS '21},
    author = {Liscio, Enrico and van der Meer, Michiel and Jonker,
    Catholijn M. and Murukannaiah, Pradeep K.},
    pages = {1773--1775},
    publisher = {IFAAMAS},
    address = {Online},
    url =
    {https://www.ifaamas.org/Proceedings/aamas2021/pdfs/p1773.pdf},
    abstract = {Value alignment is a crucial aspect of ethical
    multiagent systems. An important step toward value
    alignment is identifying values specific to an
    application context. However, identifying
    context-specific values is complex and cognitively
    demanding. To support this process, we develop a
    methodology and a collaborative web platform that
    employs AI techniques. We describe this platform,
    highlighting its intuitive design and
    implementation.}
    }

  • E. Liscio, M. van der Meer, L. C. Siebert, C. M. Jonker, N. Mouter, and P. K. Murukannaiah, “Axies: Identifying and Evaluating Context-Specific Values,” in Proc. of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021), Online, 2021, p. 799–808.
    [BibTeX] [Abstract] [Download PDF]

    The pursuit of values drives human behavior and promotes cooperation. Existing research is focused on general (e.g., Schwartz) values that transcend contexts. However, context-specific values are necessary to (1) understand human decisions, and (2) engineer intelligent agents that can elicit human values and take value-aligned actions. We propose Axies, a hybrid (human and AI) methodology to identify context-specific values. Axies simplifies the abstract task of value identification as a guided value annotation process involving human annotators. Axies exploits the growing availability of value-laden text corpora and Natural Language Processing to assist the annotators in systematically identifying context-specific values. We evaluate Axies in a user study involving 60 subjects. In our study, six annotators generate value lists for two timely and important contexts: Covid-19 measures, and sustainable Energy. Then, two policy experts and 52 crowd workers evaluate Axies value lists. We find that Axies yields values that are context-specific, consistent across different annotators, and comprehensible to end users

    @inproceedings{Liscio2021b,
    address = {Online},
    author = {Liscio, Enrico and van der Meer, Michiel and
    Siebert, Luciano C. and Jonker, Catholijn M. and
    Mouter, Niek and Murukannaiah, Pradeep K.},
    booktitle = {Proc. of the 20th International Conference on
    Autonomous Agents and Multiagent Systems (AAMAS
    2021)},
    keywords = {Context,Ethics,Natural Language
    Processing,Values,acm reference format,catholijn
    m,context,enrico liscio,ethics,jonker,luciano
    c,michiel van der meer,natural language
    processing,siebert,values},
    pages = {799--808},
    publisher = {IFAAMAS},
    title = {{Axies: Identifying and Evaluating Context-Specific
    Values}},
    year = {2021},
    url =
    "https://ii.tudelft.nl/~pradeep/doc/Liscio-2021-AAMAS-Axies.pdf",
    abstract = "The pursuit of values drives human behavior and
    promotes cooperation. Existing research is focused
    on general (e.g., Schwartz) values that transcend
    contexts. However, context-specific values are
    necessary to (1) understand human decisions, and (2)
    engineer intelligent agents that can elicit human
    values and take value-aligned actions. We propose
    Axies, a hybrid (human and AI) methodology to
    identify context-specific values. Axies simplifies
    the abstract task of value identification as a
    guided value annotation process involving human
    annotators. Axies exploits the growing availability
    of value-laden text corpora and Natural Language
    Processing to assist the annotators in
    systematically identifying context-specific values.
    We evaluate Axies in a user study involving 60
    subjects. In our study, six annotators generate
    value lists for two timely and important contexts:
    Covid-19 measures, and sustainable Energy. Then, two
    policy experts and 52 crowd workers evaluate Axies
    value lists. We find that Axies yields values that
    are context-specific, consistent across different
    annotators, and comprehensible to end users"
    }

  • K. Miras, J. Cuijpers, B. Gülhan, and A. Eiben, “The Impact of Early-death on Phenotypically Plastic Robots that Evolve in Changing Environments,” in ALIFE 2021: The 2021 Conference on Artificial Life, 2021.
    [BibTeX] [Abstract] [Download PDF]

    In this work, we evolve phenotypically plastic robots-robots that adapt their bodies and brains according to environmental conditions-in changing environments. In particular, we investigate how the possibility of death in early environmental conditions impacts evolvability and robot traits. Our results demonstrate that early-death improves the efficiency of the evolutionary process for the earlier environmental conditions. On the other hand, the possibility of early-death in the earlier environmental conditions results in a dramatic loss of performance in the latter environmental conditions.

    @inproceedings{miras2021impact,
    title = {The Impact of Early-death on Phenotypically Plastic
    Robots that Evolve in Changing Environments},
    author = {Miras, Karine and Cuijpers, Jim and Gülhan, Bahadir
    and Eiben, AE},
    booktitle = {ALIFE 2021: The 2021 Conference on Artificial Life},
    year = {2021},
    organization = {MIT Press},
    url =
    "https://direct.mit.edu/isal/proceedings-pdf/isal/33/25/1929813/isal_a_00371.pdf",
    abstract = "In this work, we evolve phenotypically plastic
    robots-robots that adapt their bodies and brains
    according to environmental conditions-in changing
    environments. In particular, we investigate how the
    possibility of death in early environmental
    conditions impacts evolvability and robot
    traits. Our results demonstrate that early-death
    improves the efficiency of the evolutionary process
    for the earlier environmental conditions. On the
    other hand, the possibility of early-death in the
    earlier environmental conditions results in a
    dramatic loss of performance in the latter
    environmental conditions."
    }

  • K. Miras, “Constrained by Design: Influence of Genetic Encodings on Evolved Traits of Robots,” Frontiers Robotics AI, vol. 8, p. 672379, 2021. doi:10.3389/frobt.2021.672379
    [BibTeX] [Abstract] [Download PDF]

    Genetic encodings and their particular properties are known to have a strong influence on the success of evolutionary systems. However, the literature has widely focused on studying the effects that encodings have on performance, i.e., fitness-oriented studies. Notably, this anchoring of the literature to performance is limiting, considering that performance provides bounded information about the behavior of a robot system. In this paper, we investigate how genetic encodings constrain the space of robot phenotypes and robot behavior. In summary, we demonstrate how two generative encodings of different nature lead to very different robots and discuss these differences. Our principal contributions are creating awareness about robot encoding biases, demonstrating how such biases affect evolved morphological, control, and behavioral traits, and finally scrutinizing the trade-offs among different biases.

    @article{DBLP:journals/firai/Miras21,
    author = {Karine Miras},
    title = {Constrained by Design: Influence of Genetic
    Encodings on Evolved Traits of Robots},
    journal = {Frontiers Robotics {AI}},
    volume = {8},
    pages = {672379},
    year = {2021},
    url = {https://doi.org/10.3389/frobt.2021.672379},
    doi = {10.3389/frobt.2021.672379},
    abstract = "Genetic encodings and their particular properties
    are known to have a strong influence on the success
    of evolutionary systems. However, the literature has
    widely focused on studying the effects that
    encodings have on performance, i.e.,
    fitness-oriented studies. Notably, this anchoring of
    the literature to performance is limiting,
    considering that performance provides bounded
    information about the behavior of a robot system. In
    this paper, we investigate how genetic encodings
    constrain the space of robot phenotypes and robot
    behavior. In summary, we demonstrate how two
    generative encodings of different nature lead to
    very different robots and discuss these
    differences. Our principal contributions are
    creating awareness about robot encoding biases,
    demonstrating how such biases affect evolved
    morphological, control, and behavioral traits, and
    finally scrutinizing the trade-offs among different
    biases."
    }

  • P. Manggala, H. H. Hoos, and E. Nalisnick, “Bayesian Regression from Multiple Sources of Weak Supervision,” in ICML 2021 Machine Learning for Data: Automated Creation, Privacy, Bias, 2021.
    [BibTeX] [Abstract] [Download PDF]

    We describe a Bayesian approach to weakly supervised regression. Our proposed framework propagates uncertainty from the weak supervision to an aggregated predictive distribution. We use a generalized Bayes procedure to account for the supervision being weak and therefore likely misspecified.

    @inproceedings{manggala2021bayesianregression,
    title = {Bayesian Regression from Multiple Sources of Weak
    Supervision},
    author = {Manggala, Putra and Hoos, Holger H. and Nalisnick,
    Eric},
    year = {2021},
    booktitle = {ICML 2021 Machine Learning for Data: Automated
    Creation, Privacy, Bias},
    url =
    {https://pmangg.github.io/papers/brfmsows_mhn_ml4data_icml.pdf},
    abstract = {We describe a Bayesian approach to weakly supervised
    regression. Our proposed framework propagates
    uncertainty from the weak supervision to an
    aggregated predictive distribution. We use a
    generalized Bayes procedure to account for the
    supervision being weak and therefore likely
    misspecified.}
    }

  • C. Steging, S. Renooij, and B. Verheij, “Discovering the rationale of decisions: towards a method for aligning learning and reasoning,” in ICAIL ’21: Eighteenth International Conference for Artificial Intelligence and Law, São Paulo Brazil, June 21 – 25, 2021, 2021, p. 235–239. doi:10.1145/3462757.3466059
    [BibTeX] [Abstract] [Download PDF]

    In AI and law, systems that are designed for decision support should be explainable when pursuing justice. In order for these systems to be fair and responsible, they should make correct decisions and make them using a sound and transparent rationale. In this paper, we introduce a knowledge-driven method for model-agnostic rationale evaluation using dedicated test cases, similar to unit-testing in professional software development. We apply this new quantitative human-in-the-loop method in a machine learning experiment aimed at extracting known knowledge structures from artificial datasets from a real-life legal setting. We show that our method allows us to analyze the rationale of black box machine learning systems by assessing which rationale elements are learned or not. Furthermore, we show that the rationale can be adjusted using tailor-made training data based on the results of the rationale evaluation.

    @inproceedings{StegingICAIL21,
    author = {Cor Steging and Silja Renooij and Bart Verheij},
    editor = {Juliano Maranh{\~{a}}o and Adam Zachary Wyner},
    title = {Discovering the rationale of decisions: towards a
    method for aligning learning and reasoning},
    booktitle = {{ICAIL} '21: Eighteenth International Conference for
    Artificial Intelligence and Law, S{\~{a}}o Paulo
    Brazil, June 21 - 25, 2021},
    pages = {235--239},
    publisher = {{ACM}},
    year = {2021},
    url = {https://doi.org/10.1145/3462757.3466059},
    doi = {10.1145/3462757.3466059},
    abstract = "In AI and law, systems that are designed for
    decision support should be explainable when pursuing
    justice. In order for these systems to be fair and
    responsible, they should make correct decisions and
    make them using a sound and transparent
    rationale. In this paper, we introduce a
    knowledge-driven method for model-agnostic rationale
    evaluation using dedicated test cases, similar to
    unit-testing in professional software
    development. We apply this new quantitative
    human-in-the-loop method in a machine learning
    experiment aimed at extracting known knowledge
    structures from artificial datasets from a real-life
    legal setting. We show that our method allows us to
    analyze the rationale of black box machine learning
    systems by assessing which rationale elements are
    learned or not. Furthermore, we show that the
    rationale can be adjusted using tailor-made training
    data based on the results of the rationale
    evaluation."
    }

  • C. Steging, S. Renooij, and B. Verheij, “Discovering the Rationale of Decisions: Experiments on Aligning Learning and Reasoning,” in 4th EXplainable AI in Law Workshop (XAILA 2021), 2021, p. 235–239.
    [BibTeX] [Abstract] [Download PDF]

    In AI and law, systems that are designed for decision support should be explainable when pursuing justice. In order for these systems to be fair and responsible, they should make correct decisions and make them using a sound and transparent rationale. In this paper, we introduce a knowledge-driven method for model-agnostic rationale evaluation using dedicated test cases, similar to unit-testing in professional software development. We apply this new method in a set of machine learning experiments aimed at extracting known knowledge structures from artificial datasets from fictional and non-fictional legal settings. We show that our method allows us to analyze the rationale of black-box machine learning systems by assessing which rationale elements are learned or not. Furthermore, we show that the rationale can be adjusted using tailor-made training data based on the results of the rationale evaluation.

    @inproceedings{StegingXAILA21,
    author = {Cor Steging and Silja Renooij and Bart Verheij},
    title = {Discovering the Rationale of Decisions: Experiments
    on Aligning Learning and Reasoning},
    maintitle = {{ICAIL} '21: Eighteenth International Conference for
    Artificial Intelligence and Law, S{\~{a}}o Paulo
    Brazil, June 21 - 25, 2021},
    booktitle = {4th EXplainable AI in Law Workshop (XAILA 2021) },
    pages = {235--239},
    publisher = {{ACM}},
    year = {2021},
    url = {https://arxiv.org/abs/2105.06758},
    abstract = "In AI and law, systems that are designed for
    decision support should be explainable when pursuing
    justice. In order for these systems to be fair and
    responsible, they should make correct decisions and
    make them using a sound and transparent
    rationale. In this paper, we introduce a
    knowledge-driven method for model-agnostic rationale
    evaluation using dedicated test cases, similar to
    unit-testing in professional software
    development. We apply this new method in a set of
    machine learning experiments aimed at extracting
    known knowledge structures from artificial datasets
    from fictional and non-fictional legal settings. We
    show that our method allows us to analyze the
    rationale of black-box machine learning systems by
    assessing which rationale elements are learned or
    not. Furthermore, we show that the rationale can be
    adjusted using tailor-made training data based on
    the results of the rationale evaluation."
    }

  • U. Khurana, E. Nalisnick, and A. Fokkens, “How Emotionally Stable is ALBERT? Testing Robustness with Stochastic Weight Averaging on a Sentiment Analysis Task,” in Proceedings of the 2nd Workshop on Evaluation and Comparison of NLP Systems, Punta Cana, Dominican Republic, 2021, p. 16–31.
    [BibTeX] [Abstract] [Download PDF]

    Despite their success, modern language models are fragile. Even small changes in their training pipeline can lead to unexpected results. We study this phenomenon by examining the robustness of ALBERT (Lan et al., 2020) in combination with Stochastic Weight Averaging (SWA){–-}a cheap way of ensembling{–-}on a sentiment analysis task (SST-2). In particular, we analyze SWA{‘}s stability via CheckList criteria (Ribeiro et al., 2020), examining the agreement on errors made by models differing only in their random seed. We hypothesize that SWA is more stable because it ensembles model snapshots taken along the gradient descent trajectory. We quantify stability by comparing the models{‘} mistakes with Fleiss{‘} Kappa (Fleiss, 1971) and overlap ratio scores. We find that SWA reduces error rates in general; yet the models still suffer from their own distinct biases (according to CheckList).

    @inproceedings{khurana-etal-2021-emotionally,
    title = "How Emotionally Stable is {ALBERT}? Testing
    Robustness with Stochastic Weight Averaging on a
    Sentiment Analysis Task",
    author = "Khurana, Urja and Nalisnick, Eric and Fokkens,
    Antske",
    booktitle = "Proceedings of the 2nd Workshop on Evaluation and
    Comparison of NLP Systems",
    month = nov,
    year = "2021",
    address = "Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.eval4nlp-1.3",
    pages = "16--31",
    abstract = "Despite their success, modern language models are
    fragile. Even small changes in their training
    pipeline can lead to unexpected results. We study
    this phenomenon by examining the robustness of
    ALBERT (Lan et al., 2020) in combination with
    Stochastic Weight Averaging (SWA){---}a cheap way of
    ensembling{---}on a sentiment analysis task
    (SST-2). In particular, we analyze SWA{'}s stability
    via CheckList criteria (Ribeiro et al., 2020),
    examining the agreement on errors made by models
    differing only in their random seed. We hypothesize
    that SWA is more stable because it ensembles model
    snapshots taken along the gradient descent
    trajectory. We quantify stability by comparing the
    models{'} mistakes with Fleiss{'} Kappa (Fleiss,
    1971) and overlap ratio scores. We find that SWA
    reduces error rates in general; yet the models still
    suffer from their own distinct biases (according to
    CheckList).",
    }

  • M. B. Vessies, S. P. Vadgama, R. van de Leur, P. A. F. M. Doevendans, R. J. Hassink, E. Bekkers, and R. Es, “Interpretable ECG classification via a query-based latent space traversal (qLST),” CoRR, vol. abs/2111.07386, 2021.
    [BibTeX] [Abstract] [Download PDF]

    Electrocardiography (ECG) is an effective and non-invasive diagnostic tool that measures the electrical activity of the heart. Interpretation of ECG signals to detect various abnormalities is a challenging task that requires expertise. Recently, the use of deep neural networks for ECG classification to aid medical practitioners has become popular, but their black box nature hampers clinical implementation. Several saliency-based interpretability techniques have been proposed, but they only indicate the location of important features and not the actual features. We present a novel interpretability technique called qLST, a query-based latent space traversal technique that is able to provide explanations for any ECG classification model. With qLST, we train a neural network that learns to traverse in the latent space of a variational autoencoder trained on a large university hospital dataset with over 800,000 ECGs annotated for 28 diseases. We demonstrate through experiments that we can explain different black box classifiers by generating ECGs through these traversals.

    @article{DBLP:journals/corr/abs-2111-07386,
    author = {Melle B. Vessies and Sharvaree P. Vadgama and Rutger
    R. van de Leur and Pieter A. F. M. Doevendans and
    Rutger J. Hassink and Erik Bekkers and Ren{\'{e}}
    van Es},
    title = {Interpretable {ECG} classification via a query-based
    latent space traversal (qLST)},
    journal = {CoRR},
    volume = {abs/2111.07386},
    year = {2021},
    url = {https://arxiv.org/abs/2111.07386},
    abstract = {Electrocardiography (ECG) is an effective and
    non-invasive diagnostic tool that measures the
    electrical activity of the heart. Interpretation of
    ECG signals to detect various abnormalities is a
    challenging task that requires expertise. Recently,
    the use of deep neural networks for ECG
    classification to aid medical practitioners has
    become popular, but their black box nature hampers
    clinical implementation. Several saliency-based
    interpretability techniques have been proposed, but
    they only indicate the location of important
    features and not the actual features. We present a
    novel interpretability technique called qLST, a
    query-based latent space traversal technique that is
    able to provide explanations for any ECG
    classification model. With qLST, we train a neural
    network that learns to traverse in the latent space
    of a variational autoencoder trained on a large
    university hospital dataset with over 800,000 ECGs
    annotated for 28 diseases. We demonstrate through
    experiments that we can explain different black box
    classifiers by generating ECGs through these
    traversals.}
    }

  • B. Dudzik and J. Broekens, “A Valid Self-Report is Never Late, Nor is it Early: On Considering the Right Temporal Distance for Assessing Emotional Experience,” , 1, 2021.
    [BibTeX] [Abstract]

    Developing computational models for automatic affect prediction requires valid self-reports about individuals’ emotional interpretations of stimuli. In this article, we highlight the important influence of the temporal distance between a stimulus event and the moment when its experience is reported on the provided information’s validity. This influence stems from the time-dependent and time-demanding nature of the involved cognitive processes. As such, reports can be collected too late: forgetting is a widely acknowledged challenge for accurate descriptions of past experience. For this reason, methods striving for assessment as early as possible have become increasingly popular. However, here we argue that collection may also occur too early: descriptions about very recent stimuli might be collected before emotional processing has fully converged. Based on these notions, we champion the existence of a temporal distance for each type of stimulus that maximizes the validity of self-reports–a” right” time. Consequently, we recommend future research to (1) consciously consider the potential influence of temporal distance on affective self-reports when planning data collection,(2) document the temporal distance of affective self-reports wherever possible as part of corpora for computational modelling, and finally (3) and explore the effect of temporal distance on self-reports across different types of stimuli.

    @techreport{Dudzik2021,
    author = {Dudzik, Bernd and Broekens, Joost},
    booktitle = {Momentary Emotion Elicitation and Capture Workshop
    (MEEC'21), May 9, 2021, Yokohama, Japan},
    number = {1},
    publisher = {Association for Computing Machinery},
    title = {{A Valid Self-Report is Never Late, Nor is it Early:
    On Considering the Right Temporal Distance for
    Assessing Emotional Experience}},
    volume = {1},
    year = {2021},
    abstract = {Developing computational models for automatic affect
    prediction requires valid self-reports about
    individuals’ emotional interpretations of
    stimuli. In this article, we highlight the important
    influence of the temporal distance between a
    stimulus event and the moment when its experience is
    reported on the provided information’s
    validity. This influence stems from the
    time-dependent and time-demanding nature of the
    involved cognitive processes. As such, reports can
    be collected too late: forgetting is a widely
    acknowledged challenge for accurate descriptions of
    past experience. For this reason, methods striving
    for assessment as early as possible have become
    increasingly popular. However, here we argue that
    collection may also occur too early: descriptions
    about very recent stimuli might be collected before
    emotional processing has fully converged. Based on
    these notions, we champion the existence of a
    temporal distance for each type of stimulus that
    maximizes the validity of self-reports–a" right"
    time. Consequently, we recommend future research to
    (1) consciously consider the potential influence of
    temporal distance on affective self-reports when
    planning data collection,(2) document the temporal
    distance of affective self-reports wherever possible
    as part of corpora for computational modelling, and
    finally (3) and explore the effect of temporal
    distance on self-reports across different types of
    stimuli.}
    }

  • B. Dudzik, S. Columbus, T. M. Hrkalovic, D. Balliet, and H. Hung, “Recognizing Perceived Interdependence in Face-to-Face Negotiations through Multimodal Analysis of Nonverbal Behavior,” in Proceedings of the 2021 International Conference on Multimodal Interaction, New York, NY, USA: Association for Computing Machinery, 2021, pp. 121-130. doi:10.1145/3462244.3479935
    [BibTeX] [Abstract] [Download PDF]

    Enabling computer-based applications to display intelligent behavior in complex social settings requires them to relate to important aspects of how humans experience and understand such situations. One crucial driver of peoples’ social behavior during an interaction is the interdependence they perceive, i.e., how the outcome of an interaction is determined by their own and others’ actions. According to psychological studies, both the nonverbal behavior displayed by Motivated by this, we present a series of experiments to automatically recognize interdependence perceptions in dyadic face-to-face negotiations using these sources. Concretely, our approach draws on a combination of features describing individuals’ Facial, Upper Body, and Vocal Behavior with state-of-the-art algorithms for multivariate time series classification. Our findings demonstrate that differences in some types of interdependence perceptions can be detected through the automatic analysis of nonverbal behaviors. We discuss implications for developing socially intelligent systems and opportunities for future research.

    @inbook{10.1145/3462244.3479935,
    author = {Dudzik, Bernd and Columbus, Simon and Hrkalovic,
    Tiffany Matej and Balliet, Daniel and Hung, Hayley},
    title = {Recognizing Perceived Interdependence in
    Face-to-Face Negotiations through Multimodal
    Analysis of Nonverbal Behavior},
    year = {2021},
    isbn = {9781450384810},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    booktitle = {Proceedings of the 2021 International Conference on
    Multimodal Interaction},
    pages = {121-130},
    numpages = {10},
    doi = {10.1145/3462244.3479935},
    url =
    {https://research.tudelft.nl/en/publications/recognizing-perceived-interdependence-in-face-to-face-negotiation},
    abstract = {Enabling computer-based applications to display
    intelligent behavior in complex social settings
    requires them to relate to important aspects of how
    humans experience and understand such
    situations. One crucial driver of peoples' social
    behavior during an interaction is the
    interdependence they perceive, i.e., how the outcome
    of an interaction is determined by their own and
    others' actions. According to psychological studies,
    both the nonverbal behavior displayed by Motivated
    by this, we present a series of experiments to
    automatically recognize interdependence perceptions
    in dyadic face-to-face negotiations using these
    sources. Concretely, our approach draws on a
    combination of features describing individuals'
    Facial, Upper Body, and Vocal Behavior with
    state-of-the-art algorithms for multivariate time
    series classification. Our findings demonstrate that
    differences in some types of interdependence
    perceptions can be detected through the automatic
    analysis of nonverbal behaviors. We discuss
    implications for developing socially intelligent
    systems and opportunities for future research.}
    }

  • C. Steging, S. Renooij, and B. Verheij, “Rationale Discovery and Explainable AI,” in Legal Knowledge and Information Systems – JURIX 2021: The Thirty-fourth Annual Conference, Vilnius, Lithuania, 8-10 December 2021, 2021, p. 225–234. doi:10.3233/FAIA210341
    [BibTeX] [Abstract] [Download PDF]

    The justification of an algorithm’s outcomes is important in many domains, and in particular in the law. However, previous research has shown that machine learning systems can make the right decisions for the wrong reasons: despite high accuracies, not all of the conditions that define the domain of the training data are learned. In this study, we investigate what the system does learn, using state-of-the-art explainable AI techniques. With the use of SHAP and LIME, we are able to show which features impact the decision making process and how the impact changes with different distributions of the training data. However, our results also show that even high accuracy and good relevant feature detection are no guarantee for a sound rationale. Hence these state-of-the-art explainable AI techniques cannot be used to fully expose unsound rationales, further advocating the need for a separate method for rationale evaluation.

    @inproceedings{DBLP:conf/jurix/StegingRV21,
    author = {Cor Steging and Silja Renooij and Bart Verheij},
    editor = {Schweighofer Erich},
    title = {Rationale Discovery and Explainable {AI}},
    booktitle = {Legal Knowledge and Information Systems - {JURIX}
    2021: The Thirty-fourth Annual Conference, Vilnius,
    Lithuania, 8-10 December 2021},
    series = {Frontiers in Artificial Intelligence and
    Applications},
    volume = {346},
    pages = {225--234},
    publisher = {{IOS} Press},
    year = {2021},
    url = {https://doi.org/10.3233/FAIA210341},
    doi = {10.3233/FAIA210341},
    abstract = {The justification of an algorithm’s
    outcomes is important in many domains, and in
    particular in the law. However, previous research
    has shown that machine learning systems can make the
    right decisions for the wrong reasons: despite high
    accuracies, not all of the conditions that define
    the domain of the training data are learned. In this
    study, we investigate what the system does learn,
    using state-of-the-art explainable AI
    techniques. With the use of SHAP and LIME, we are
    able to show which features impact the decision
    making process and how the impact changes with
    different distributions of the training
    data. However, our results also show that even high
    accuracy and good relevant feature detection are no
    guarantee for a sound rationale. Hence these
    state-of-the-art explainable AI techniques cannot be
    used to fully expose unsound rationales, further
    advocating the need for a separate method for
    rationale evaluation.}
    }

  • H. Loan, “Knowledge Representation Formalisms for Hybrid Intelligence,” in Doctoral consortium of the Knowledge Representation Conference, KR2021, online, 2021.
    [BibTeX] [Abstract] [Download PDF]

    Knowledge graphs can play an important role to store and provide access to global knowledge, common and accessible to both human and artificial agents, and store local knowledge of individual agents in a larger network of agents. Studying suitable formalisms to model complex, conflicting, dynamic and contextualised knowledge is still a big challenge. Therefore, we investigate the usage of knowledge representation formalisms to allow artificial intelligence systems adapt and work with complex, conflicting, dynamic and contextualized knowledge.

    @inproceedings{Loan2021,
    address = {online},
    author = {Loan, Ho},
    publisher = {KR Conference},
    title = {Knowledge Representation Formalisms for Hybrid
    Intelligence},
    year = {2021},
    booktitle = {Doctoral consortium of the Knowledge Representation Conference, {KR}2021},
    url = "https://sites.google.com/view/kr2021dc",
    abstract = "Knowledge graphs can play an important role to store
    and provide access to global knowledge, common and
    accessible to both human and artificial agents, and
    store local knowledge of individual agents in a
    larger network of agents. Studying suitable
    formalisms to model complex, conflicting, dynamic
    and contextualised knowledge is still a big
    challenge. Therefore, we investigate the usage of
    knowledge representation formalisms to allow
    artificial intelligence systems adapt and work with
    complex, conflicting, dynamic and contextualized
    knowledge."
    }

  • B. H. Kargar, K. Miras, and A. Eiben, “The effect of selecting for different behavioral traits on the evolved gaits of modular robots,” in ALIFE 2021: The 2021 Conference on Artificial Life, 2021.
    [BibTeX] [Abstract] [Download PDF]

    Moving around in the environment is a fundamental skill for mobile robots. This makes the evolution of an appropriate gait, a pivotal problem in evolutionary robotics. Whereas the majority of the related studies concern robots with predefined modular or legged morphologies and locomotion speed as the optimization objective, here we investigate robots with evolvable morphologies and behavioral traits included in the fitness function. To analyze the effects we consider morphological as well as behavioral features of the evolved robots. To this end, we introduce novel behavioral measures that describe how the robot locomotes and look into the trade-off between them. Our main goal is to gain insights into differences in possible gaits of modular robots and to provide tools to steer evolution towards objectives beyond ‘simple’ speed.

    @inproceedings{kargar2021effect,
    title = {The effect of selecting for different behavioral
    traits on the evolved gaits of modular robots},
    author = {Kargar, Babak H and Miras, Karine and Eiben, AE},
    booktitle = {ALIFE 2021: The 2021 Conference on Artificial Life},
    year = {2021},
    organization = {MIT Press},
    url =
    {https://direct.mit.edu/isal/proceedings/isal/33/26/102968},
    abstract = {Moving around in the environment is a fundamental
    skill for mobile robots. This makes the evolution of
    an appropriate gait, a pivotal problem in
    evolutionary robotics. Whereas the majority of the
    related studies concern robots with predefined
    modular or legged morphologies and locomotion speed
    as the optimization objective, here we investigate
    robots with evolvable morphologies and behavioral
    traits included in the fitness function. To analyze
    the effects we consider morphological as well as
    behavioral features of the evolved robots. To this
    end, we introduce novel behavioral measures that
    describe how the robot locomotes and look into the
    trade-off between them. Our main goal is to gain
    insights into differences in possible gaits of
    modular robots and to provide tools to steer
    evolution towards objectives beyond 'simple' speed.}
    }

  • G. Boomgaard, S. B. Santamaría, I. Tiddi, R. J. Sips, and Z. Szávik, “Learning profile-based recommendations for medical search auto-complete,” in AAAI-MAKE 2021 Combining Machine Learning and Knowledge Engineering, 2021, p. 1–13.
    [BibTeX] [Abstract] [Download PDF]

    Query popularity is a main feature in web-search auto-completion. Several personalization features have been proposed to support specific users’ searches, but often do not meet the privacy requirements of a medical environment (e.g. clinical trial search). Furthermore, in such specialized domains, the differences in user expertise and the domain-specific language users employ are far more widespread than in web-search. We propose a query auto-completion method based on different relevancy and diversity features, which can appropriately meet different user needs. Our method incorporates indirect popularity measures, along with graph topology and semantic features. An evolutionary algorithm optimizes relevance, diversity, and coverage to return a top-k list of query completions to the user. We evaluated our approach quantitatively and qualitatively using query log data from a clinical trial search engine, comparing the effects of different relevancy and diversity settings using domain experts. We found that syntax-based diversity has more impact on effectiveness and efficiency, graph-based diversity shows a more compact list of results, and relevancy the most effect on indicated preferences.

    @inproceedings{boomgaard-etal-2021-learning,
    title = "Learning profile-based recommendations for medical
    search auto-complete",
    author = "Guusje Boomgaard and Selene Baez Santamaría and Ilaria Tiddi and Robert Jan Sips
    and Zoltán Szávik",
    keywords = "Knowledge graphs, Medical information retrieval,
    Professional search, Query auto-Completion",
    year = "2021",
    month = apr,
    day = "10",
    language = "English",
    series = "CEUR Workshop Proceedings",
    publisher = "CEUR-WS",
    pages = "1--13",
    editor = "Andreas Martin and Knut Hinkelmann and Hans-Georg
    Fill and Aurona Gerber and Doug Lenat and Reinhard
    Stolle and {van Harmelen}, Frank",
    booktitle = "AAAI-MAKE 2021 Combining Machine Learning and
    Knowledge Engineering",
    Url = "http://ceur-ws.org/Vol-2846/paper34.pdf",
    abstract = "Query popularity is a main feature in web-search
    auto-completion. Several personalization features
    have been proposed to support specific users'
    searches, but often do not meet the privacy
    requirements of a medical environment (e.g. clinical
    trial search). Furthermore, in such specialized
    domains, the differences in user expertise and the
    domain-specific language users employ are far more
    widespread than in web-search. We propose a query
    auto-completion method based on different relevancy
    and diversity features, which can appropriately meet
    different user needs. Our method incorporates
    indirect popularity measures, along with graph
    topology and semantic features. An evolutionary
    algorithm optimizes relevance, diversity, and
    coverage to return a top-k list of query completions
    to the user. We evaluated our approach
    quantitatively and qualitatively using query log
    data from a clinical trial search engine, comparing
    the effects of different relevancy and diversity
    settings using domain experts. We found that
    syntax-based diversity has more impact on
    effectiveness and efficiency, graph-based diversity
    shows a more compact list of results, and relevancy
    the most effect on indicated preferences.",
    }

  • S. Baez Santamaria, T. Baier, T. Kim, L. Krause, J. Kruijt, and P. Vossen, “EMISSOR: A platform for capturing multimodal interactions as Episodic Memories and Interpretations with Situated Scenario-based Ontological References,” in Proceedings of the 1st Workshop on Multimodal Semantic Representations (MMSR), Groningen, Netherlands (Online), 2021, p. 56–77.
    [BibTeX] [Abstract] [Download PDF]

    We present EMISSOR: a platform to capture multimodal interactions as recordings of episodic experiences with explicit referential interpretations that also yield an episodic Knowledge Graph (eKG). The platform stores streams of multiple modalities as parallel signals. Each signal is segmented and annotated independently with interpretation. Annotations are eventually mapped to explicit identities and relations in the eKG. As we ground signal segments from different modalities to the same instance representations, we also ground different modalities across each other. Unique to our eKG is that it accepts different interpretations across modalities, sources and experiences and supports reasoning over conflicting information and uncertainties that may result from multimodal experiences. EMISSOR can record and annotate experiments in virtual and real-world, combine data, evaluate system behavior and their performance for preset goals but also model the accumulation of knowledge and interpretations in the Knowledge Graph as a result of these episodic experiences.

    @inproceedings{baez-santamaria-etal-2021-emissor,
    title = "{EMISSOR}: A platform for capturing multimodal
    interactions as Episodic Memories and
    Interpretations with Situated Scenario-based
    Ontological References",
    author = "Baez Santamaria, Selene and Baier, Thomas and Kim,
    Taewoon and Krause, Lea and Kruijt, Jaap and Vossen,
    Piek",
    booktitle = "Proceedings of the 1st Workshop on Multimodal
    Semantic Representations (MMSR)",
    month = jun,
    year = "2021",
    address = "Groningen, Netherlands (Online)",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.mmsr-1.6",
    pages = "56--77",
    abstract = "We present EMISSOR: a platform to capture multimodal
    interactions as recordings of episodic experiences
    with explicit referential interpretations that also
    yield an episodic Knowledge Graph (eKG). The
    platform stores streams of multiple modalities as
    parallel signals. Each signal is segmented and
    annotated independently with
    interpretation. Annotations are eventually mapped to
    explicit identities and relations in the eKG. As we
    ground signal segments from different modalities to
    the same instance representations, we also ground
    different modalities across each other. Unique to
    our eKG is that it accepts different interpretations
    across modalities, sources and experiences and
    supports reasoning over conflicting information and
    uncertainties that may result from multimodal
    experiences. EMISSOR can record and annotate
    experiments in virtual and real-world, combine data,
    evaluate system behavior and their performance for
    preset goals but also model the accumulation of
    knowledge and interpretations in the Knowledge Graph
    as a result of these episodic experiences.",
    }

  • R. Dobbe, T. Krendl Gilbert, and Y. Mintz, “Hard choices in artificial intelligence,” Artificial Intelligence, vol. 300, 2021. doi:10.1016/j.artint.2021.103555
    [BibTeX] [Abstract] [Download PDF]

    As AI systems are integrated into high stakes social domains, researchers now examine how to design and operate them in a safe and ethical manner. However, the criteria for identifying and diagnosing safety risks in complex social contexts remain unclear and contested. In this paper, we examine the vagueness in debates about the safety and ethical behavior of AI systems. We show how this vagueness cannot be resolved through mathematical formalism alone, instead requiring deliberation about the politics of development as well as the context of deployment. Drawing from a new sociotechnical lexicon, we redefine vagueness in terms of distinct design challenges at key stages in AI system development. The resulting framework of Hard Choices in Artificial Intelligence (HCAI) empowers developers by 1. identifying points of overlap between design decisions and major sociotechnical challenges; 2. motivating the creation of stakeholder feedback channels so that safety issues can be exhaustively addressed. As such, HCAI contributes to a timely debate about the status of AI development in democratic societies, arguing that deliberation should be the goal of AI Safety, not just the procedure by which it is ensured.

    @article{dobbe_hard_2021,
    title = {Hard choices in artificial intelligence},
    volume = {300},
    issn = {0004-3702},
    url =
    {https://www.sciencedirect.com/science/article/pii/S0004370221001065},
    doi = {10.1016/j.artint.2021.103555},
    abstract = {As AI systems are integrated into high stakes social
    domains, researchers now examine how to design and
    operate them in a safe and ethical manner. However,
    the criteria for identifying and diagnosing safety
    risks in complex social contexts remain unclear and
    contested. In this paper, we examine the vagueness
    in debates about the safety and ethical behavior of
    AI systems. We show how this vagueness cannot be
    resolved through mathematical formalism alone,
    instead requiring deliberation about the politics of
    development as well as the context of
    deployment. Drawing from a new sociotechnical
    lexicon, we redefine vagueness in terms of distinct
    design challenges at key stages in AI system
    development. The resulting framework of Hard Choices
    in Artificial Intelligence (HCAI) empowers
    developers by 1. identifying points of overlap
    between design decisions and major sociotechnical
    challenges; 2. motivating the creation of
    stakeholder feedback channels so that safety issues
    can be exhaustively addressed. As such, HCAI
    contributes to a timely debate about the status of
    AI development in democratic societies, arguing that
    deliberation should be the goal of AI Safety, not
    just the procedure by which it is ensured.},
    language = {en},
    urldate = {2021-08-04},
    journal = {Artificial Intelligence},
    author = {Dobbe, Roel and Krendl Gilbert, Thomas and Mintz,
    Yonatan},
    month = nov,
    year = {2021},
    keywords = {AI ethics, AI governance, AI regulation, AI safety,
    Philosophy of artificial intelligence,
    Sociotechnical systems}
    }

  • T. Koopman and S. Renooij, “Persuasive Contrastive Explanations for Bayesian Networks,” in Proceedings of the Sixteenth European Conference on Symbolic and Quantitative Approached to Reasoning with Uncertainty (ECSQARU), 2021, p. 229–242. doi:10.1007/978-3-030-86772-0\{_}17
    [BibTeX] [Abstract] [Download PDF]

    Explanation in Artificial Intelligence is often focused on providing reasons for why a model under consideration and its outcome are correct. Recently, research in explainable machine learning has initiated a shift in focus on including so-called counterfactual explanations. In this paper we propose to combine both types of explanation in the context of explaining Bayesian networks. To this end we introduce persuasive contrastive explanations that aim to provide an answer to the question Why outcome t instead of t’? posed by a user. In addition, we propose an algorithm for computing persuasive contrastive explanations. Both our definition of persuasive contrastive explanation and the proposed algorithm can be employed beyond the current scope of Bayesian networks.

    @InProceedings{ecsqaru-koopman21,
    title = {Persuasive Contrastive Explanations for {B}ayesian
    Networks},
    author = {Koopman, Tara AND Renooij, Silja},
    booktitle = { Proceedings of the Sixteenth European Conference on
    Symbolic and Quantitative Approached to Reasoning
    with Uncertainty (ECSQARU)},
    year = {2021},
    pages = {229--242},
    editor = {Vejnarova, J. and Wilson, N.},
    volume = {12897},
    series = {Lecture Notes in Computer Science},
    month = {21--24 Sept},
    publisher = {Springer, Cham},
    pdf =
    {https://webspace.science.uu.nl/~renoo101/Prof/PDF/Conf/ecsqaru2021-final.pdf},
    doi = {10.1007/978-3-030-86772-0\{_}17},
    abstract = {Explanation in Artificial Intelligence is often
    focused on providing reasons for why a model under
    consideration and its outcome are correct. Recently,
    research in explainable machine learning has
    initiated a shift in focus on including so-called
    counterfactual explanations. In this paper we
    propose to combine both types of explanation in the
    context of explaining Bayesian networks. To this end
    we introduce persuasive contrastive explanations
    that aim to provide an answer to the question Why
    outcome t instead of t'? posed by a user. In
    addition, we propose an algorithm for computing
    persuasive contrastive explanations. Both our
    definition of persuasive contrastive explanation and
    the proposed algorithm can be employed beyond the
    current scope of Bayesian networks.},
    URL =
    {https://webspace.science.uu.nl/~renoo101/Prof/PDF/Conf/ecsqaru2021-final.pdf}
    }

  • T. Koopman and S. Renooij, “Persuasive Contrastive Explanations (Extended Abstract),” in Proceedings of The 2nd Workshop in Explainable Logic-based Knowledge Representation (XLoKR), 2021.
    [BibTeX] [Abstract] [Download PDF]

    Explanation in Artificial Intelligence is often focused on providing reasons for why a model under consideration and its outcome are correct. Recently, research in explainable machine learning has initiated a shift in focus on including so-called counterfactual explanations. In this paper we propose to combine both types of explanation into a persuasive contrastive explanation that aims to provide an answer to the question Why outcome t instead of t’? posed by a user. In addition, we propose a model-agnostic algorithm for computing persuasive contrastive explanations from AI systems with few input variables.

    @InProceedings{xlokr-koopman21,
    title = {Persuasive Contrastive Explanations (Extended
    Abstract)},
    author = {Koopman, Tara AND Renooij, Silja},
    booktitle = {Proceedings of The 2nd Workshop in Explainable
    Logic-based Knowledge Representation (XLoKR)},
    year = {2021},
    editor = {Baader, F. AND Bogaerts, B. AND Brewka, G. AND
    Hoffmann, J. AND Lukasiewicz, T. AND Potyka, N. AND
    Toni, F.},
    month = {04--05 NOV},
    pdf = {https://xlokr21.ai.vub.ac.be/papers/16/paper.pdf},
    abstract = {Explanation in Artificial Intelligence is often
    focused on providing reasons for why a model under
    consideration and its outcome are correct. Recently,
    research in explainable machine learning has
    initiated a shift in focus on including so-called
    counterfactual explanations. In this paper we
    propose to combine both types of explanation into a
    persuasive contrastive explanation that aims to
    provide an answer to the question Why outcome t
    instead of t'? posed by a user. In addition, we
    propose a model-agnostic algorithm for computing
    persuasive contrastive explanations from AI systems
    with few input variables.},
    URL = {https://xlokr21.ai.vub.ac.be/papers/16/paper.pdf},
    }

  • M. A. Rahman, N. Hopner, F. Christianos, and S. V. Albrecht, “Towards Open Ad Hoc Teamwork Using Graph-based Policy Learning,” in Proceedings of the 38th International Conference on Machine Learning, 2021, p. 8776–8786.
    [BibTeX] [Abstract] [Download PDF]

    Ad hoc teamwork is the challenging problem of designing an autonomous agent which can adapt quickly to collaborate with teammates without prior coordination mechanisms, including joint training. Prior work in this area has focused on closed teams in which the number of agents is fixed. In this work, we consider open teams by allowing agents with different fixed policies to enter and leave the environment without prior notification. Our solution builds on graph neural networks to learn agent models and joint-action value models under varying team compositions. We contribute a novel action-value computation that integrates the agent model and joint-action value model to produce action-value estimates. We empirically demonstrate that our approach successfully models the effects other agents have on the learner, leading to policies that robustly adapt to dynamic team compositions and significantly outperform several alternative methods.

    @InProceedings{pmlr-v139-rahman21a,
    title = {Towards Open Ad Hoc Teamwork Using Graph-based
    Policy Learning},
    author = {Rahman, Muhammad A and Hopner, Niklas and
    Christianos, Filippos and Albrecht, Stefano V},
    booktitle = {Proceedings of the 38th International Conference on
    Machine Learning},
    pages = {8776--8786},
    year = {2021},
    editor = {Meila, Marina and Zhang, Tong},
    volume = {139},
    series = {Proceedings of Machine Learning Research},
    month = {18--24 Jul},
    publisher = {PMLR},
    pdf =
    {http://proceedings.mlr.press/v139/rahman21a/rahman21a.pdf},
    url = {https://proceedings.mlr.press/v139/rahman21a.html},
    abstract = {Ad hoc teamwork is the challenging problem of
    designing an autonomous agent which can adapt
    quickly to collaborate with teammates without prior
    coordination mechanisms, including joint
    training. Prior work in this area has focused on
    closed teams in which the number of agents is
    fixed. In this work, we consider open teams by
    allowing agents with different fixed policies to
    enter and leave the environment without prior
    notification. Our solution builds on graph neural
    networks to learn agent models and joint-action
    value models under varying team compositions. We
    contribute a novel action-value computation that
    integrates the agent model and joint-action value
    model to produce action-value estimates. We
    empirically demonstrate that our approach
    successfully models the effects other agents have on
    the learner, leading to policies that robustly adapt
    to dynamic team compositions and significantly
    outperform several alternative methods.}
    }

2020

  • B. Verheij, “Artificial intelligence as law,” Artif. Intell. Law, vol. 28, iss. 2, p. 181–206, 2020. doi:10.1007/s10506-020-09266-0
    [BibTeX] [Abstract] [Download PDF]

    Information technology is so ubiquitous and AI’s progress so inspiring that also legal professionals experience its benefits and have high expectations. At the same time, the powers of AI have been rising so strongly that it is no longer obvious that AI applications (whether in the law or elsewhere) help promoting a good society; in fact they are sometimes harmful. Hence many argue that safeguards are needed for AI to be trustworthy, social, responsible, humane, ethical. In short: AI should be good for us. But how to establish proper safeguards for AI? One strong answer readily available is: consider the problems and solutions studied in AI & Law. AI & Law has worked on the design of social, explainable, responsible AI aligned with human values for decades already, AI & Law addresses the hardest problems across the breadth of AI (in reasoning, knowledge, learning and language), and AI & Law inspires new solutions (argumentation, schemes and norms, rules and cases, interpretation). It is argued that the study of AI as Law supports the development of an AI that is good for us, making AI & Law more relevant than ever.

    @article{Verheij20,
    author = {Bart Verheij},
    title = {Artificial intelligence as law},
    journal = {Artif. Intell. Law},
    volume = {28},
    number = {2},
    pages = {181--206},
    year = {2020},
    url = {https://doi.org/10.1007/s10506-020-09266-0},
    doi = {10.1007/s10506-020-09266-0},
    timestamp = {Fri, 05 Jun 2020 17:08:42 +0200},
    biburl = {https://dblp.org/rec/journals/ail/Verheij20.bib},
    bibsource = {dblp computer science bibliography,
    https://dblp.org},
    abstract = {Information technology is so ubiquitous and AI’s
    progress so inspiring that also legal professionals
    experience its benefits and have high
    expectations. At the same time, the powers of AI
    have been rising so strongly that it is no longer
    obvious that AI applications (whether in the law or
    elsewhere) help promoting a good society; in fact
    they are sometimes harmful. Hence many argue that
    safeguards are needed for AI to be trustworthy,
    social, responsible, humane, ethical. In short: AI
    should be good for us. But how to establish proper
    safeguards for AI? One strong answer readily
    available is: consider the problems and solutions
    studied in AI & Law. AI & Law has worked on the
    design of social, explainable, responsible AI
    aligned with human values for decades already, AI &
    Law addresses the hardest problems across the
    breadth of AI (in reasoning, knowledge, learning and
    language), and AI & Law inspires new solutions
    (argumentation, schemes and norms, rules and cases,
    interpretation). It is argued that the study of AI
    as Law supports the development of an AI that is
    good for us, making AI & Law more relevant than
    ever.}
    }

  • N. Kökciyan and P. Yolum, “TURP: Managing Trust for Regulating Privacy in Internet of Things,” IEEE Internet Computing, vol. 24, iss. 6, pp. 9-16, 2020. doi:https://doi.org/10.1109/MIC.2020.3020006
    [BibTeX] [Abstract] [Download PDF]

    Internet of Things [IoT] applications, such as smart home or ambient assisted livingsystems, promise useful services to end users. Most of these services rely heavily on sharingand aggregating information among devices; many times raising privacy concerns. Contrary totraditional systems, where privacy of each user is managed through well-defined policies, thescale, dynamism, and heterogeneity of the IoT systems make it impossible to specify privacypolicies for all possible situations. Alternatively, this paper argues that handling of privacy has tobe reasoned by the IoT devices, depending on the norms, context, as well as the trust amongentities. We present a technique, where an IoT device collects information from others, evaluatesthe trustworthiness of the information sources to decide the suitability of sharing informationwith others. We demonstrate the applicability of the technique over an IoT pilot study.

    @ARTICLE{turp-ic-2020,
    author = {K\"okciyan, Nadin and Yolum, P{\i}nar},
    journal = {IEEE Internet Computing},
    title = {TURP: Managing Trust for Regulating Privacy in
    Internet of Things},
    year = {2020},
    volume = {24},
    number = {6},
    pages = {9-16},
    abstract = {Internet of Things [IoT] applications, such as smart
    home or ambient assisted livingsystems, promise
    useful services to end users. Most of these services
    rely heavily on sharingand aggregating information
    among devices; many times raising privacy
    concerns. Contrary totraditional systems, where
    privacy of each user is managed through well-defined
    policies, thescale, dynamism, and heterogeneity of
    the IoT systems make it impossible to specify
    privacypolicies for all possible
    situations. Alternatively, this paper argues that
    handling of privacy has tobe reasoned by the IoT
    devices, depending on the norms, context, as well as
    the trust amongentities. We present a technique,
    where an IoT device collects information from
    others, evaluatesthe trustworthiness of the
    information sources to decide the suitability of
    sharing informationwith others. We demonstrate the
    applicability of the technique over an IoT pilot
    study.},
    url =
    {https://webspace.science.uu.nl/~yolum001/papers/InternetComputing-20-TURP.pdf},
    doi = {https://doi.org/10.1109/MIC.2020.3020006}
    }

  • O. Ulusoy and P. Yolum, “Agents for Preserving Privacy: Learning and Decision Making Collaboratively,” in Multi-Agent Systems and Agreement Technologies, 2020, p. 116–131. doi:https://doi.org/10.1007/978-3-030-66412-1_8
    [BibTeX] [Abstract] [Download PDF]

    Privacy is a right of individuals to keep personal information to themselves. Often online systems enable their users to select what information they would like to share with others and what information to keep private. When an information pertains only to a single individual, it is possible to preserve privacy by providing the right access options to the user. However, when an information pertains to multiple individuals, such as a picture of a group of friends or a collaboratively edited document, deciding how to share this information and with whom is challenging as individuals might have conflicting privacy constraints. Resolving this problem requires an automated mechanism that takes into account the relevant individuals’ concerns to decide on the privacy configuration of information. Accordingly, this paper proposes an auction-based privacy mechanism to manage the privacy of users when information related to multiple individuals are at stake. We propose to have a software agent that acts on behalf of each user to enter privacy auctions, learn the subjective privacy valuations of the individuals over time, and to bid to respect their privacy. We show the workings of our proposed approach over multiagent simulations.

    @InProceedings{ulusoy-yolum-20,
    title = "Agents for Preserving Privacy: Learning and Decision
    Making Collaboratively",
    author = "Ulusoy, Onuralp and Yolum, P{\i}nar",
    editor = "Bassiliades, Nick and Chalkiadakis, Georgios and de
    Jonge, Dave",
    booktitle = "Multi-Agent Systems and Agreement Technologies",
    year = "2020",
    publisher = "Springer International Publishing",
    pages = "116--131",
    abstract = "Privacy is a right of individuals to keep personal
    information to themselves. Often online systems
    enable their users to select what information they
    would like to share with others and what information
    to keep private. When an information pertains only
    to a single individual, it is possible to preserve
    privacy by providing the right access options to the
    user. However, when an information pertains to
    multiple individuals, such as a picture of a group
    of friends or a collaboratively edited document,
    deciding how to share this information and with whom
    is challenging as individuals might have conflicting
    privacy constraints. Resolving this problem requires
    an automated mechanism that takes into account the
    relevant individuals' concerns to decide on the
    privacy configuration of information. Accordingly,
    this paper proposes an auction-based privacy
    mechanism to manage the privacy of users when
    information related to multiple individuals are at
    stake. We propose to have a software agent that acts
    on behalf of each user to enter privacy auctions,
    learn the subjective privacy valuations of the
    individuals over time, and to bid to respect their
    privacy. We show the workings of our proposed
    approach over multiagent simulations.",
    isbn = "978-3-030-66412-1",
    doi = {https://doi.org/10.1007/978-3-030-66412-1_8},
    url =
    {https://webspace.science.uu.nl/~yolum001/papers/ulusoy-yolum-20.pdf}
    }

  • L. Krause and P. Vossen, “When to explain: Identifying explanation triggers in human-agent interaction,” in 2nd Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence, Dublin, Ireland, 2020, p. 55–60.
    [BibTeX] [Abstract] [Download PDF]

    With more agents deployed than ever, users need to be able to interact and cooperate with them in an effective and comfortable manner. Explanations have been shown to increase the understanding and trust of a user in human-agent interaction. There have been numerous studies investigating this effect, but they rely on the user explicitly requesting an explanation. We propose a first overview of when an explanation should be triggered and show that there are many instances that would be missed if the agent solely relies on direct questions. For this, we differentiate between direct triggers such as commands or questions and introduce indirect triggers like confusion or uncertainty detection.

    @inproceedings{krause-vossen-2020-explain,
    title = "When to explain: Identifying explanation triggers in
    human-agent interaction",
    author = "Krause, Lea and Vossen, Piek",
    booktitle = "2nd Workshop on Interactive Natural Language
    Technology for Explainable Artificial Intelligence",
    month = nov,
    year = "2020",
    address = "Dublin, Ireland",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.nl4xai-1.12",
    pages = "55--60",
    abstract = "With more agents deployed than ever, users need to
    be able to interact and cooperate with them in an
    effective and comfortable manner. Explanations have
    been shown to increase the understanding and trust
    of a user in human-agent interaction. There have
    been numerous studies investigating this effect, but
    they rely on the user explicitly requesting an
    explanation. We propose a first overview of when an
    explanation should be triggered and show that there
    are many instances that would be missed if the agent
    solely relies on direct questions. For this, we
    differentiate between direct triggers such as
    commands or questions and introduce indirect
    triggers like confusion or uncertainty detection.",
    }

  • P. K. Murukannaiah, N. Ajmeri, C. M. Jonker, and M. P. Singh, “New Foundations of Ethical Multiagent Systems,” in Proceedings of the 19th Conference on Autonomous Agents and MultiAgent Systems, Auckland, 2020, p. 1706–1710.
    [BibTeX] [Abstract] [Download PDF]

    Ethics is inherently a multiagent concern. However, research on AI ethics today is dominated by work on individual agents: (1) how an autonomous robot or car may harm or (differentially) benefit people in hypothetical situations (the so-called trolley problems) and (2) how a machine learning algorithm may produce biased decisions or recommendations. The societal framework is largely omitted. To develop new foundations for ethics in AI, we adopt a sociotechnical stance in which agents (as technical entities) help autonomous social entities or principals (people and organizations). This multiagent conception of a sociotechnical system (STS) captures how ethical concerns arise in the mutual interactions of multiple stakeholders. These foundations would enable us to realize ethical STSs that incorporate social and technical controls to respect stated ethical postures of the agents in the STSs. The envisioned foundations require new thinking, along two broad themes, on how to realize (1) an STS that reflects its stakeholders’ values and (2) individual agents that function effectively in such an STS.

    @inproceedings{Murukannaiah-2020-AAMASBlueSky-EthicalMAS,
    author = {Pradeep K. Murukannaiah and Nirav Ajmeri and
    Catholijn M. Jonker and Munindar P. Singh},
    title = {New Foundations of Ethical Multiagent Systems},
    booktitle = {Proceedings of the 19th Conference on Autonomous
    Agents and MultiAgent Systems},
    series = {AAMAS '20},
    year = {2020},
    address = {Auckland},
    pages = {1706--1710},
    numpages = {5},
    keywords = {Agents, ethics},
    url =
    {https://ii.tudelft.nl/~pradeep/doc/Murukannaiah-2020-AAMASBlueSky-EthicalMAS.pdf},
    abstract = {Ethics is inherently a multiagent concern. However,
    research on AI ethics today is dominated by work on
    individual agents: (1) how an autonomous robot or
    car may harm or (differentially) benefit people in
    hypothetical situations (the so-called trolley
    problems) and (2) how a machine learning algorithm
    may produce biased decisions or recommendations. The
    societal framework is largely omitted. To develop
    new foundations for ethics in AI, we adopt a
    sociotechnical stance in which agents (as technical
    entities) help autonomous social entities or
    principals (people and organizations). This
    multiagent conception of a sociotechnical system
    (STS) captures how ethical concerns arise in the
    mutual interactions of multiple stakeholders. These
    foundations would enable us to realize ethical STSs
    that incorporate social and technical controls to
    respect stated ethical postures of the agents in the
    STSs. The envisioned foundations require new
    thinking, along two broad themes, on how to realize
    (1) an STS that reflects its stakeholders' values
    and (2) individual agents that function effectively
    in such an STS.}
    }

  • Z. Akata, D. Balliet, M. de Rijke, F. Dignum, V. Dignum, G. Eiben, A. Fokkens, D. Grossi, K. Hindriks, H. Hoos, H. Hung, C. Jonker, C. Monz, M. Neerincx, F. Oliehoek, H. Prakken, S. Schlobach, L. van der Gaag, F. van Harmelen, H. van Hoof, B. van Riemsdijk, A. van Wynsberghe, R. Verbrugge, B. Verheij, P. Vossen, and M. Welling, “A Research Agenda for Hybrid Intelligence: Augmenting Human Intellect With Collaborative, Adaptive, Responsible, and Explainable Artificial Intelligence,” IEEE Computer, vol. 53, iss. 08, pp. 18-28, 2020. doi:10.1109/MC.2020.2996587
    [BibTeX] [Abstract] [Download PDF]

    We define hybrid intelligence (HI) as the combination of human and machine intelligence, augmenting human intellect and capabilities instead of replacing them and achieving goals that were unreachable by either humans or machines. HI is an important new research focus for artificial intelligence, and we set a research agenda for HI by formulating four challenges.

    @ARTICLE {9153877,
    author = {Z. Akata and D. Balliet and M. de Rijke and
    F. Dignum and V. Dignum and G. Eiben and A. Fokkens
    and D. Grossi and K. Hindriks and H. Hoos and
    H. Hung and C. Jonker and C. Monz and M. Neerincx
    and F. Oliehoek and H. Prakken and S. Schlobach and
    L. van der Gaag and F. van Harmelen and H. van Hoof
    and B. van Riemsdijk and A. van Wynsberghe and
    R. Verbrugge and B. Verheij and P. Vossen and
    M. Welling},
    journal = {IEEE Computer},
    title = {A Research Agenda for Hybrid Intelligence:
    Augmenting Human Intellect With Collaborative,
    Adaptive, Responsible, and Explainable Artificial
    Intelligence},
    year = {2020},
    volume = {53},
    number = {08},
    issn = {1558-0814},
    pages = {18-28},
    doi = {10.1109/MC.2020.2996587},
    publisher = {IEEE Computer Society},
    address = {Los Alamitos, CA, USA},
    month = {aug},
    url =
    "http://www.cs.vu.nl/~frankh/postscript/IEEEComputer2020.pdf",
    abstract = "We define hybrid intelligence (HI) as the
    combination of human and machine intelligence,
    augmenting human intellect and capabilities instead
    of replacing them and achieving goals that were
    unreachable by either humans or machines. HI is an
    important new research focus for artificial
    intelligence, and we set a research agenda for HI by
    formulating four challenges."
    }

  • B. M. Renting, H. H. Hoos, and C. M. Jonker, “Automated Configuration of Negotiation Strategies,” in Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems, 2020, p. 1116–1124.
    [BibTeX] [Abstract] [Download PDF]

    Bidding and acceptance strategies have a substantial impact on the outcome of negotiations in scenarios with linear additive and nonlinear utility functions. Over the years, it has become clear that there is no single best strategy for all negotiation settings, yet many fixed strategies are still being developed. We envision a shift in the strategy design question from: What is a good strategy?, towards: What could be a good strategy? For this purpose, we developed a method leveraging automated algorithm configuration to find the best strategies for a specific set of negotiation settings. By empowering automated negotiating agents using automated algorithm configuration, we obtain a flexible negotiation agent that can be configured automatically for a rich space of opponents and negotiation scenarios. To critically assess our approach, the agent was tested in an ANAC-like bilateral automated negotiation tournament setting against past competitors. We show that our automatically configured agent outperforms all other agents, with a 5.1 percent increase in negotiation payoff compared to the next-best agent. We note that without our agent in the tournament, the top-ranked agent wins by a margin of only 0.01 percent .

    @inproceedings{Renting2020AutomatedStrategies,
    title = {Automated Configuration of Negotiation Strategies},
    booktitle = {Proceedings of the 19th International Conference on
    Autonomous Agents and MultiAgent Systems},
    author = {Renting, Bram M. and Hoos, Holger H. and Jonker,
    Catholijn M.},
    year = {2020},
    month = {may},
    series = {AAMAS '20},
    pages = {1116--1124},
    publisher = {International Foundation for Autonomous Agents and
    Multiagent Systems},
    abstract = {Bidding and acceptance strategies have a substantial
    impact on the outcome of negotiations in scenarios
    with linear additive and nonlinear utility
    functions. Over the years, it has become clear that
    there is no single best strategy for all negotiation
    settings, yet many fixed strategies are still being
    developed. We envision a shift in the strategy
    design question from: What is a good strategy?,
    towards: What could be a good strategy? For this
    purpose, we developed a method leveraging automated
    algorithm configuration to find the best strategies
    for a specific set of negotiation settings. By
    empowering automated negotiating agents using
    automated algorithm configuration, we obtain a
    flexible negotiation agent that can be configured
    automatically for a rich space of opponents and
    negotiation scenarios. To critically assess our
    approach, the agent was tested in an ANAC-like
    bilateral automated negotiation tournament setting
    against past competitors. We show that our
    automatically configured agent outperforms all other
    agents, with a 5.1 percent increase in negotiation payoff
    compared to the next-best agent. We note that
    without our agent in the tournament, the top-ranked
    agent wins by a margin of only 0.01 percent .},
    isbn = {978-1-4503-7518-4},
    keywords = {automated algorithm configuration,automated
    negotiation,negotiation strategy},
    url = {https://ifaamas.org/Proceedings/aamas2020/pdfs/p1116.pdf}
    }

  • B. M. Renting, H. H. Hoos, and C. M. Jonker, “Automated Configuration and Usage of Strategy Portfolios for Bargaining.” 2021-12-14. doi:10.48550/arXiv.2212.10228
    [BibTeX] [Abstract] [Download PDF]

    Bargaining can be used to resolve mixed-motive games in multi-agent systems. Although there is an abundance of negotiation strategies implemented in automated negotiating agents, most agents are based on single fixed strategies, while it is widely acknowledged that there is no single best-performing strategy for all negotiation settings. In this paper, we focus on bargaining settings where opponents are repeatedly encountered, but the bargaining problems change. We introduce a novel method that automatically creates and deploys a portfolio of complementary negotiation strategies using a training set and optimise pay-off in never-before-seen bargaining settings through per-setting strategy selection. Our method relies on the following contributions. We introduce a feature representation that captures characteristics for both the opponent and the bargaining problem. We model the behaviour of an opponent during a negotiation based on its actions, which is indicative of its negotiation strategy, in order to be more effective in future encounters. Our combination of feature-based methods generalises to new negotiation settings, as in practice, over time, it selects effective counter strategies in future encounters. Our approach is tested in an ANAC-like tournament, and we show that we are capable of winning such a tournament with a 5.6 percent increase in pay-off compared to the runner-up agent.

    @inproceedings{rentingAutomatedConfigurationUsage2021,
    title = {Automated Configuration and Usage of Strategy Portfolios for Bargaining},
    author = {Renting, Bram M. and Hoos, Holger H. and Jonker, Catholijn M.},
    date = {2021-12-14},
    eprint = {2212.10228},
    eprinttype = {arxiv},
    primaryclass = {cs},
    publisher = {{arXiv}},
    doi = {10.48550/arXiv.2212.10228},
    url = {http://arxiv.org/abs/2212.10228},
    urldate = {2022-12-29},
    abstract = {Bargaining can be used to resolve mixed-motive games in multi-agent systems. Although there is an abundance of negotiation strategies implemented in automated negotiating agents, most agents are based on single fixed strategies, while it is widely acknowledged that there is no single best-performing strategy for all negotiation settings. In this paper, we focus on bargaining settings where opponents are repeatedly encountered, but the bargaining problems change. We introduce a novel method that automatically creates and deploys a portfolio of complementary negotiation strategies using a training set and optimise pay-off in never-before-seen bargaining settings through per-setting strategy selection. Our method relies on the following contributions. We introduce a feature representation that captures characteristics for both the opponent and the bargaining problem. We model the behaviour of an opponent during a negotiation based on its actions, which is indicative of its negotiation strategy, in order to be more effective in future encounters. Our combination of feature-based methods generalises to new negotiation settings, as in practice, over time, it selects effective counter strategies in future encounters. Our approach is tested in an ANAC-like tournament, and we show that we are capable of winning such a tournament with a 5.6 percent increase in pay-off compared to the runner-up agent.},
    archiveprefix = {arXiv},
    eventtitle = {The {{Second Cooperative AI}} Workshop, {{NeurIPS}} 2021},
    keywords = {Computer Science - Multiagent Systems},
    }

  • B. M. Renting, H. H. Hoos, and C. M. Jonker, “Automated Configuration and Usage of Strategy Portfolios for Mixed-Motive Bargaining,” in Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems, 2022-05-09, p. 1101–1109.
    [BibTeX] [Abstract] [Download PDF]

    Bargaining can be used to resolve mixed-motive games in multi-agent systems. Although there is an abundance of negotiation strategies implemented in automated negotiating agents, most agents are based on single fixed strategies, while it is acknowledged that there is no single best-performing strategy for all negotiation settings. In this paper, we focus on bargaining settings where opponents are repeatedly encountered, but the bargaining problems change. We introduce a novel method that automatically creates and deploys a portfolio of complementary negotiation strategies using a training set and optimise pay-off in never-before-seen bargaining settings through per-setting strategy selection. Our method relies on the following contributions. We introduce a feature representation that captures characteristics for both the opponent and the bargaining problem. We model the behaviour of an opponent during a negotiation based on its actions, which is indicative of its negotiation strategy, in order to be more effective in future encounters. Our combination of feature-based methods generalises to new negotiation settings, as in practice, over time, it selects effective counter strategies in future encounters. Our approach is tested in an ANAC-like tournament, and we show that we are capable of winning such a tournament with a 5.6 percent increase in pay-off compared to the runner-up agent.

    @inproceedings{rentingAutomatedConfigurationUsage2022,
    title = {Automated Configuration and Usage of Strategy Portfolios for Mixed-Motive Bargaining},
    booktitle = {Proceedings of the 21st {{International Conference}} on {{Autonomous Agents}} and {{Multiagent Systems}}},
    author = {Renting, Bram M. and Hoos, Holger H. and Jonker, Catholijn M.},
    date = {2022-05-09},
    series = {{{AAMAS}} '22},
    pages = {1101--1109},
    publisher = {{International Foundation for Autonomous Agents and Multiagent Systems}},
    location = {{Richland, SC}},
    url = {https://ifaamas.org/Proceedings/aamas2022/pdfs/p1101.pdf},
    urldate = {2022-10-23},
    abstract = {Bargaining can be used to resolve mixed-motive games in multi-agent systems. Although there is an abundance of negotiation strategies implemented in automated negotiating agents, most agents are based on single fixed strategies, while it is acknowledged that there is no single best-performing strategy for all negotiation settings. In this paper, we focus on bargaining settings where opponents are repeatedly encountered, but the bargaining problems change. We introduce a novel method that automatically creates and deploys a portfolio of complementary negotiation strategies using a training set and optimise pay-off in never-before-seen bargaining settings through per-setting strategy selection. Our method relies on the following contributions. We introduce a feature representation that captures characteristics for both the opponent and the bargaining problem. We model the behaviour of an opponent during a negotiation based on its actions, which is indicative of its negotiation strategy, in order to be more effective in future encounters. Our combination of feature-based methods generalises to new negotiation settings, as in practice, over time, it selects effective counter strategies in future encounters. Our approach is tested in an ANAC-like tournament, and we show that we are capable of winning such a tournament with a 5.6 percent increase in pay-off compared to the runner-up agent.},
    isbn = {978-1-4503-9213-6},
    keywords = {algorithm configuration,algorithm selection,bargaining,mixed-motive games},
    }