Nachtegael, Charlotte; Stefani, Jacopo De; Cnudde, Anthony; Lenaerts, Tom
DUVEL: an active-learning annotated biomedical corpus for the recognition of oligogenic combinations Journal Article
In: Database, vol. 2024, no. 2024, 2024, (DOI: 10.1093/database/baae039).
@article{info:hdl:2013/374632b,
title = {DUVEL: an active-learning annotated biomedical corpus for the recognition of oligogenic combinations},
author = {Charlotte Nachtegael and Jacopo De Stefani and Anthony Cnudde and Tom Lenaerts},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/374632/1/doi_358276.pdf},
year = {2024},
date = {2024-01-01},
journal = {Database},
volume = {2024},
number = {2024},
abstract = {Abstract While biomedical relation extraction (bioRE) datasets have been instrumental in the development of methods to support biocuration of single variants from texts, no datasets are currently available for the extraction of digenic or even oligogenic variant relations, despite the reports in literature that epistatic effects between combinations of variants in different loci (or genes) are important to understand disease etiologies. This work presents the creation of a unique dataset of oligogenic variant combinations, geared to train tools to help in the curation of scientific literature. To overcome the hurdles associated with the number of unlabelled instances and the cost of expertise, active learning (AL) was used to optimize the annotation, thus getting assistance in finding the most informative subset of samples to label. By pre-annotating 85 full-text articles containing the relevant relations from the Oligogenic Diseases Database (OLIDA) with PubTator, text fragments featuring potential digenic variant combinations, i.e. gene–variant–gene–variant, were extracted. The resulting fragments of texts were annotated with ALAMBIC, an AL-based annotation platform. The resulting dataset, called DUVEL, is used to fine-tune four state-of-the-art biomedical language models: BiomedBERT, BiomedBERT-large, BioLinkBERT and BioM-BERT. More than 500 000 text fragments were considered for annotation, finally resulting in a dataset with 8442 fragments, 794 of them being positive instances, covering 95% of the original annotated articles. When applied to gene–variant pair detection, BiomedBERT-large achieves the highest F1 score (0.84) after fine-tuning, demonstrating significant improvement compared to the non-fine-tuned model, underlining the relevance of the DUVEL dataset. This study shows how AL may play an important role in the creation of bioRE dataset relevant for biomedical curation applications. DUVEL provides a unique biomedical corpus focusing on 4-ary relations between two genes and two variants. It is made freely available for research on GitHub and Hugging Face. Database URL: https://huggingface.co/datasets/cnachteg/duvel or https://doi.org/10.57967/hf/1571},
note = {DOI: 10.1093/database/baae039},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Kirchsteiger, Georg; Lenaerts, Tom; Suchon, Remi
Voluntary versus mandatory information disclosure in the sequential prisoner’s dilemma Journal Article
In: Economic theory, 2024, (DOI: 10.1007/s00199-024-01563-y).
@article{info:hdl:2013/373750b,
title = {Voluntary versus mandatory information disclosure in the sequential prisoner’s dilemma},
author = {Georg Kirchsteiger and Tom Lenaerts and Remi Suchon},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/373750/3/s00199-024-01563-y.pdf},
year = {2024},
date = {2024-01-01},
journal = {Economic theory},
abstract = {In sequential social dilemmas with stranger matching, initiating cooperation is inherently risky for the first mover. The disclosure of the second mover’s past actions may be necessary to instigate cooperation. We experimentally compare the effect of mandatory and voluntary disclosure with non-disclosure in a sequential prisoner’s dilemma situation. Our results confirm the positive effects of disclosure on cooperation. We also find that voluntary disclosure is as effective as mandatory disclosure, which runs counter to the results of existing literature on this topic. With voluntary disclosure, second movers who have a good track record chose to disclose, suggesting that they anticipate non-disclosure would signal non-cooperativeness. First movers interpret non-disclosure correctly as a signal of non-cooperativeness. Therefore, they cooperate less than half as often when the second mover decides not to disclose.},
note = {DOI: 10.1007/s00199-024-01563-y},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Terrucha, Ines; Domingos, Elias Fernández; Simoens, Pieter; Lenaerts, Tom
Committing to the wrong artificial delegate in a collective-risk dilemma is better than directly committing mistakes Journal Article
In: Scientific reports, vol. 14, no. 1, 2024, (DOI: 10.1038/s41598-024-61153-9).
@article{info:hdl:2013/374814b,
title = {Committing to the wrong artificial delegate in a collective-risk dilemma is better than directly committing mistakes},
author = {Ines Terrucha and Elias Fernández Domingos and Pieter Simoens and Tom Lenaerts},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/374814/1/doi_358458.pdf},
year = {2024},
date = {2024-01-01},
journal = {Scientific reports},
volume = {14},
number = {1},
abstract = {While autonomous artificial agents are assumed to perfectly execute the strategies they are programmed with, humans who design them may make mistakes. These mistakes may lead to a misalignment between the humans’ intended goals and their agents’ observed behavior, a problem of value alignment. Such an alignment problem may have particularly strong consequences when these autonomous systems are used in social contexts that involve some form of collective risk. By means of an evolutionary game theoretical model, we investigate whether errors in the configuration of artificial agents change the outcome of a collective-risk dilemma, in comparison to a scenario with no delegation. Delegation is here distinguished from no-delegation simply by the moment at which a mistake occurs: either when programming/choosing the agent (in case of delegation) or when executing the actions at each round of the game (in case of no-delegation). We find that, while errors decrease success rate, it is better to delegate and commit to a somewhat flawed strategy, perfectly executed by an autonomous agent, than to commit execution errors directly. Our model also shows that in the long-term, delegation strategies should be favored over no-delegation, if given the choice.},
note = {DOI: 10.1038/s41598-024-61153-9},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Rivière, Quentin; Raskin, Virginie; Melo, Romário; Boutet, Stéphanie; Corso, Massimiliano; Defrance, Matthieu; Webb, Alex A. R.; Verbruggen, Nathalie; Anoman, Djoro Armand
Effects of light regimes on circadian gene co‐expression networks in Arabidopsis thaliana Journal Article
In: Plant Direct, vol. 8, no. 8, 2024, (DOI: 10.1002/pld3.70001).
@article{info:hdl:2013/384388b,
title = {Effects of light regimes on circadian gene co‐expression networks in Arabidopsis thaliana},
author = {Quentin Rivière and Virginie Raskin and Romário Melo and Stéphanie Boutet and Massimiliano Corso and Matthieu Defrance and Alex A. R. Webb and Nathalie Verbruggen and Djoro Armand Anoman},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/384388/3/Plant-Direct-2024.pdf},
year = {2024},
date = {2024-01-01},
journal = {Plant Direct},
volume = {8},
number = {8},
abstract = {Abstract Light/dark (LD) cycles are responsible for oscillations in gene expression, which modulate several aspects of plant physiology. Those oscillations can persist under constant conditions due to regulation by the circadian oscillator. The response of the transcriptome to light regimes is dynamic and allows plants to adapt rapidly to changing environmental conditions. We compared the transcriptome of Arabidopsis under LD and constant light (LL) for 3 days and identified different gene co‐expression networks in the two light regimes. Our studies yielded unforeseen insights into circadian regulation. Intuitively, we anticipated that gene clusters regulated by the circadian oscillator would display oscillations under LD cycles. However, we found transcripts encoding components of the flavonoid metabolism pathway that were rhythmic in LL but not in LD. We also discovered that the expressions of many stress‐related genes were significantly increased during the dark period in LD relative to the subjective night in LL, whereas the expression of these genes in the light period was similar. The nocturnal pattern of these stress‐related gene expressions suggested a form of “skotoprotection.” The transcriptomics data were made available in a web application named Cyclath , which we believe will be a useful tool to contribute to a better understanding of the impact of light regimes on plants.},
note = {DOI: 10.1002/pld3.70001},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Jansen, Maarten
Information criteria for structured parameter selection in high dimensional tree and graph models Journal Article
In: Digital signal processing, vol. 148, 2024, (Language of publication: fr).
@article{info:hdl:2013/372845b,
title = {Information criteria for structured parameter selection in high dimensional tree and graph models},
author = {Maarten Jansen},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/372845/3/jansen24structuredpreprint.pdf},
year = {2024},
date = {2024-01-01},
journal = {Digital signal processing},
volume = {148},
note = {Language of publication: fr},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Bhattacharya, Shreya; Lefèvre, Laure; Chatzistergos, T; Hayakawa, Hisashi; Jansen, Maarten
RudolfWolf to AlfredWolfer: The Transfer of the Reference Observer in the International Sunspot Number Series (1876–1893) Journal Article
In: Solar physics, vol. 299, 2024, (Language of publication: fr).
@article{info:hdl:2013/372844b,
title = {RudolfWolf to AlfredWolfer: The Transfer of the Reference Observer in the International Sunspot Number Series (1876–1893)},
author = {Shreya Bhattacharya and Laure Lefèvre and T Chatzistergos and Hisashi Hayakawa and Maarten Jansen},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/372844/3/PreprintBhattacharya202404.pdf},
year = {2024},
date = {2024-01-01},
journal = {Solar physics},
volume = {299},
note = {Language of publication: fr},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Jansen, Maarten; Claeskens, Gerda
The Cramér-Rao Lower Bound Book Chapter
In: 2024, (Language of publication: fr).
@inbook{info:hdl:2013/359778b,
title = {The Cramér-Rao Lower Bound},
author = {Maarten Jansen and Gerda Claeskens},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/359778/3/jansen23cramerraoencycl.pdf},
year = {2024},
date = {2024-01-01},
note = {Language of publication: fr},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Jansen, Maarten; Claeskens, G.
Nonparametric estimation Book Chapter
In: 2024, (Language of publication: fr).
@inbook{info:hdl:2013/359775b,
title = {Nonparametric estimation},
author = {Maarten Jansen and G. Claeskens},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/359775/3/jansen23nonparametricencycl.pdf},
year = {2024},
date = {2024-01-01},
note = {Language of publication: fr},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Colot, Martin; Simar, Cédric; Petieau, Mathieu; Alvarez, Ana Maria Cebolla; Chéron, Guy; Bontempi, Gianluca
EMG subspace alignment and visualization for cross-subject hand gesture classification Miscellaneous
2024, (Conference: ECML-PKDD 2023 Worshop - Adapting to change : Reliable Learning Across Domains (2023-09-18: Turin)).
@misc{info:hdl:2013/373864b,
title = {EMG subspace alignment and visualization for cross-subject hand gesture classification},
author = {Martin Colot and Cédric Simar and Mathieu Petieau and Ana Maria Cebolla Alvarez and Guy Chéron and Gianluca Bontempi},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/373864/3/ECML_PKDD_2023__Martin_Colot_Main_and_Appendix.pdf},
year = {2024},
date = {2024-01-01},
abstract = {Electromyograms (EMG)-based hand gesture recognition systems are a promising technology for human/machine interfaces. However, one of their main limitations is the long calibration time that is typically required to handle new users. The paper discusses and analyses the challenge of cross-subject generalization thanks to an original dataset containing the EMG signals of 14 human subjects during hand gestures. The experimental results show that, though an accurate generalization based on pooling multiple subjects is hardly achievable, it is possible to improve the cross-subject estimation by identifying a robust low-dimensional subspace for multiple subjects and aligning it to a target subject. A visualization of the subspace enables us to provide insights for the improvement of cross-subject generalization with EMG signals.},
note = {Conference: ECML-PKDD 2023 Worshop - Adapting to change : Reliable Learning Across Domains (2023-09-18: Turin)},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Lebichot, Bertrand; Siblini, Wissam; Paldino, Gian Marco; Borgne, Yann-Aël Le; Oblé, Frédéric; Bontempi, Gianluca
Assessment of catastrophic forgetting in continual credit card fraud detection Journal Article
In: Expert systems with applications, vol. 249, 2024, (DOI: 10.1016/j.eswa.2024.123445).
@article{info:hdl:2013/370795b,
title = {Assessment of catastrophic forgetting in continual credit card fraud detection},
author = {Bertrand Lebichot and Wissam Siblini and Gian Marco Paldino and Yann-Aël Le Borgne and Frédéric Oblé and Gianluca Bontempi},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/370795/3/preproofs-main.pdf},
year = {2024},
date = {2024-01-01},
journal = {Expert systems with applications},
volume = {249},
abstract = {The volume of e-commerce continues to increase year after year. Buying goods on the internet is easy and practical, and took a huge boost during the lockdowns of the Covid crisis. However, this is also an open window for fraudsters and the corresponding financial loss costs billions of dollars. In this paper, we study e-commerce credit card fraud detection, in collaboration with our industrial partner, Worldline. Transactional companies are more and more dependent on machine learning models such as deep learning anomaly detection models, as part of real-world fraud detection systems (FDS). We focus on continual learning to find the best model with respect to two objectives: to maximize the accuracy and to minimize the catastrophic forgetting phenomenon. For the latter, we proposed an evaluation procedure to quantify the forgetting in data streams with delayed feedback: the plasticity/stability visualization matrix. We also investigated six strategies and 13 methods on a real-size case study including five months of e-commerce credit card transactions. Finally, we discuss how the trade-off between plasticity and stability is set, in practice, in the case of FDS.},
note = {DOI: 10.1016/j.eswa.2024.123445},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Paldino, Gian Marco; Lebichot, Bertrand; Borgne, Yann-Aël Le; Siblini, Wissam; Oblé, Frédéric; Boracchi, Giacomo; Bontempi, Gianluca
The role of diversity and ensemble learning in credit card fraud detection Journal Article
In: Advances in Data Analysis and Classification, vol. 18, no. 1, pp. 193-217, 2024, (DOI: 10.1007/s11634-022-00515-5).
@article{info:hdl:2013/372242b,
title = {The role of diversity and ensemble learning in credit card fraud detection},
author = {Gian Marco Paldino and Bertrand Lebichot and Yann-Aël Le Borgne and Wissam Siblini and Frédéric Oblé and Giacomo Boracchi and Gianluca Bontempi},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/372242/1/elsevier_355886.pdf},
year = {2024},
date = {2024-01-01},
journal = {Advances in Data Analysis and Classification},
volume = {18},
number = {1},
pages = {193-217},
abstract = {The number of daily credit card transactions is inexorably growing: the e-commerce market expansion and the recent constraints for the Covid-19 pandemic have significantly increased the use of electronic payments. The ability to precisely detect fraudulent transactions is increasingly important, and machine learning models are now a key component of the detection process. Standard machine learning techniques are widely employed, but inadequate for the evolving nature of customers behavior entailing continuous changes in the underlying data distribution. his problem is often tackled by discarding past knowledge, despite its potential relevance in the case of recurrent concepts. Appropriate exploitation of historical knowledge is necessary: we propose a learning strategy that relies on diversity-based ensemble learning and allows to preserve past concepts and reuse them for a faster adaptation to changes. In our experiments, we adopt several state-of-the-art diversity measures and we perform comparisons with various other learning approaches. We assess the effectiveness of our proposed learning strategy on extracts of two real datasets from two European countries, containing more than 30 M and 50 M transactions, provided by our industrial partner, Worldline, a leading company in the field.},
note = {DOI: 10.1007/s11634-022-00515-5},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Simar, Cédric; Colot, Martin; Alvarez, Ana Maria Cebolla; Petieau, Mathieu; Chéron, Guy; Bontempi, Gianluca
Machine learning for hand pose classification from phasic and tonic EMG signals during bimanual activities in virtual reality Journal Article
In: Frontiers in Neuroscience, vol. 18, 2024, (DOI: 10.3389/fnins.2024.1329411).
@article{info:hdl:2013/373455b,
title = {Machine learning for hand pose classification from phasic and tonic EMG signals during bimanual activities in virtual reality},
author = {Cédric Simar and Martin Colot and Ana Maria Cebolla Alvarez and Mathieu Petieau and Guy Chéron and Gianluca Bontempi},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/373455/1/doi_357099.pdf},
year = {2024},
date = {2024-01-01},
journal = {Frontiers in Neuroscience},
volume = {18},
abstract = {Myoelectric prostheses have recently shown significant promise for restoring hand function in individuals with upper limb loss or deficiencies, driven by advances in machine learning and increasingly accessible bioelectrical signal acquisition devices. Here, we first introduce and validate a novel experimental paradigm using a virtual reality headset equipped with hand-tracking capabilities to facilitate the recordings of synchronized EMG signals and hand pose estimation. Using both the phasic and tonic EMG components of data acquired through the proposed paradigm, we compare hand gesture classification pipelines based on standard signal processing features, convolutional neural networks, and covariance matrices with Riemannian geometry computed from raw or xDAWN-filtered EMG signals. We demonstrate the performance of the latter for gesture classification using EMG signals. We further hypothesize that introducing physiological knowledge in machine learning models will enhance their performances, leading to better myoelectric prosthesis control. We demonstrate the potential of this approach by using the neurophysiological integration of the “move command" to better separate the phasic and tonic components of the EMG signals, significantly improving the performance of sustained posture recognition. These results pave the way for the development of new cutting-edge machine learning techniques, likely refined by neurophysiology, that will further improve the decoding of real-time natural gestures and, ultimately, the control of myoelectric prostheses.},
note = {DOI: 10.3389/fnins.2024.1329411},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Cerqueira, Vitor; Torgo, Luis; Bontempi, Gianluca
Instance-based meta-learning for conditionally dependent univariate multi-step forecasting Journal Article
In: International journal of forecasting, 2024, (DOI: 10.1016/j.ijforecast.2023.12.010).
@article{info:hdl:2013/371938b,
title = {Instance-based meta-learning for conditionally dependent univariate multi-step forecasting},
author = {Vitor Cerqueira and Luis Torgo and Gianluca Bontempi},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/371938/3/ijf.meta2.pdf},
year = {2024},
date = {2024-01-01},
journal = {International journal of forecasting},
abstract = {Multi-step prediction is a key challenge in univariate forecasting. However, forecasting accuracy decreases as predictions are made further into the future. This is caused by the decreasing predictability and the error propagation along the horizon. In this paper, we propose a novel method called Forecasted Trajectory Neighbors (FTN) for multi-step forecasting with univariate time series. FTN is a meta-learning strategy that can be integrated with any state-of-the-art multi-step forecasting approach. It works by using training observations to correct the errors made during multiple predictions. This is accomplished by retrieving the nearest neighbors of the multi-step forecasts and averaging these for prediction. The motivation is to introduce, in a lightweight manner, a conditional dependent constraint across the forecasting horizons. Such a constraint, not always taken into account by most strategies, can be considered as a sort of regularization element. We carried out extensive experiments using 7795 time series from different application domains. We found that our method improves the performance of several state-of-the-art multi-step forecasting methods. An implementation of the proposed method is publicly available online, and the experiments are reproducible.},
note = {DOI: 10.1016/j.ijforecast.2023.12.010},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Terrucha, Ines; Domingos, Elias Fernandez; Suchon, Remi; Santos, Francisco C; Simoens, Pieter; Lenaerts, Tom
Humans program artificial delegates to accurately solve collective-risk dilemmas, but lack precision Miscellaneous
2024, (Conference: Machine+behavior Conference(Berlin, Allemagne)).
@misc{info:hdl:2013/385912,
title = {Humans program artificial delegates to accurately solve collective-risk dilemmas, but lack precision},
author = {Ines Terrucha and Elias Fernandez Domingos and Remi Suchon and Francisco C Santos and Pieter Simoens and Tom Lenaerts},
url = {http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/385912},
year = {2024},
date = {2024-01-01},
note = {Conference: Machine+behavior Conference(Berlin, Allemagne)},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Kirchsteiger, Georg; Lenaerts, Tom; Suchon, Remi
Growing cooperation Miscellaneous
2024, (Conference: Conference of the French Experimental Economics Association(14: grenoble, France)).
@misc{info:hdl:2013/385911,
title = {Growing cooperation},
author = {Georg Kirchsteiger and Tom Lenaerts and Remi Suchon},
url = {http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/385911},
year = {2024},
date = {2024-01-01},
note = {Conference: Conference of the French Experimental Economics Association(14: grenoble, France)},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Bosch, Inas; Gravel, Barbara; Lenaerts, Tom
Knowledge graph embeddings for the prediction of pathogenic gene pairs Miscellaneous
2024, (Conference: European Conference on Computational Biology.(23: 16/09-20/09/2024: Turku, Finland)).
@misc{info:hdl:2013/385910,
title = {Knowledge graph embeddings for the prediction of pathogenic gene pairs},
author = {Inas Bosch and Barbara Gravel and Tom Lenaerts},
url = {http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/385910},
year = {2024},
date = {2024-01-01},
note = {Conference: European Conference on Computational Biology.(23: 16/09-20/09/2024: Turku, Finland)},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Gravel, Barbara; Renaux, Alexandre; Papadimitriou, Sofia; Smits, Guillaume; Nowé, Ann; Lenaerts, Tom
Prioritization of variant combinations in whole exomes Miscellaneous
2024, (Conference: European Conference on Computational Biology.(23: 16/09-20/09/2024: Turku, Finland)).
@misc{info:hdl:2013/385909,
title = {Prioritization of variant combinations in whole exomes},
author = {Barbara Gravel and Alexandre Renaux and Sofia Papadimitriou and Guillaume Smits and Ann Nowé and Tom Lenaerts},
url = {http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/385909},
year = {2024},
date = {2024-01-01},
note = {Conference: European Conference on Computational Biology.(23: 16/09-20/09/2024: Turku, Finland)},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Abels, Axel; Lenaerts, Tom; Trianni, Vito; Nowé, Ann
Dealing with Expert Bias in Collective Decision-making Miscellaneous
2024, (Conference: European Conference on Artificial Intelligence(27: 19/10-24/10/2024: Santiago de Compostella)).
@misc{info:hdl:2013/385908,
title = {Dealing with Expert Bias in Collective Decision-making},
author = {Axel Abels and Tom Lenaerts and Vito Trianni and Ann Nowé},
url = {http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/385908},
year = {2024},
date = {2024-01-01},
note = {Conference: European Conference on Artificial Intelligence(27: 19/10-24/10/2024: Santiago de Compostella)},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Leung, Chin Wing; Lenaerts, Tom; Turrini, Paolo
To Promote Full Cooperation in Social Dilemmas, Agents Need to Unlearn Loyalty Proceedings Article
In: Larson, Kate (Ed.): Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, pp. 111-119, International Joint Conferences on Artificial Intelligence (IJCAI) Organization, 2024, (Conference: International Joint Conference on Artificial Intelligence(33: 3/8-9/8/2024: Jeju. Korea)).
@inproceedings{info:hdl:2013/385907,
title = {To Promote Full Cooperation in Social Dilemmas, Agents Need to Unlearn Loyalty},
author = {Chin Wing Leung and Tom Lenaerts and Paolo Turrini},
editor = {Kate Larson},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/385907/3/0013.pdf},
year = {2024},
date = {2024-01-01},
booktitle = {Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence},
pages = {111-119},
publisher = {International Joint Conferences on Artificial Intelligence (IJCAI) Organization},
abstract = {If given the choice, what strategy should agents use to switch partners in strategic social interactions? While many analyses have been performed on specific switching heuristics, showing how and when these lead to more cooperation, no insights have been provided into which rule will actually be learnt by agents when given the freedom to do so. Starting from a baseline model that has demonstrated the potential of rewiring for cooperation, we provide answers to this question over the full spectrum of social dilemmas. Multi-agent Q-learning with Boltzmann exploration is used to learn when to sever or maintain an association. In both the Prisoner's Dilemma and the Stag Hunt games we observe that the Out-for-Tat rewiring rule, breaking ties with other agents choosing socially undesirable actions, becomes dominant, confirming at the same time that cooperation flourishes when rewiring is fast enough relative to imitation. Nonetheless, in the transitory region before full cooperation, a Stay strategy, keeping a connection at all costs, remains present, which shows that loyalty needs to be overcome for full cooperation to emerge. In conclusion, individuals learn cooperation-promoting rewiring rules but need to overcome a kind of loyalty to achieve full cooperation in the full spectrum of social dilemmas.},
note = {Conference: International Joint Conference on Artificial Intelligence(33: 3/8-9/8/2024: Jeju. Korea)},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Molinghen, Yannick; Avalos, Raphaël; Achter, Mark Van; Nowé, Ann; Lenaerts, Tom
Laser Learning Environment: A new environment for coordination-critical multi-agent tasks Proceedings Article
In: Oliehoek, Frans F. A.; Manon, Kok; Verwer, Sicco (Ed.): Artificial Intelligence and Machine Learning: Revised Selected Papers, Springer Science and Business Media Deutschland GmbH, 2024, (Conference: Benelux Conference Ai conference, BNAIC(35: 8-10/11/2023: TU Delft)).
@inproceedings{info:hdl:2013/370546,
title = {Laser Learning Environment: A new environment for coordination-critical multi-agent tasks},
author = {Yannick Molinghen and Raphaël Avalos and Mark Van Achter and Ann Nowé and Tom Lenaerts},
editor = {Frans F. A. Oliehoek and Kok Manon and Sicco Verwer},
url = {https://dipot.ulb.ac.be/dspace/bitstream/2013/370546/4/2404.03596v1.pdf},
year = {2024},
date = {2024-01-01},
booktitle = {Artificial Intelligence and Machine Learning: Revised Selected Papers},
publisher = {Springer Science and Business Media Deutschland GmbH},
series = {Communications in Computer and Information Science},
abstract = {We introduce the Laser Learning Environment (LLE), a collaborative multi-agent reinforcement learning environment where coordination is key. In LLE, agents depend on each other to make progress (interdependence), must jointly take specific sequences of actions to succeed (perfect coordination), and accomplishing those joint actions does not yield any intermediate reward (zero-incentive dynamics). The challenge of such problems lies in the difficulty of escaping state space bottlenecks caused by interdependence steps since escaping those bottlenecks is not rewarded. We test multiple state-of-the-art value-based MARL algorithms against LLE and show that they consistently fail at the collaborative task because of their inability to escape state space bottlenecks, even though they successfully achieve perfect coordination. We show that Q-learning extensions such as prioritised experience replay and n-steps return hinder exploration in environments with zero-incentive dynamics, and find that intrinsic curiosity with random network distillation is not sufficient to escape those bottlenecks. We demonstrate the need for novel methods to solve this problem and the relevance of LLE as cooperative MARL benchmark.},
note = {Conference: Benelux Conference Ai conference, BNAIC(35: 8-10/11/2023: TU Delft)},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}