Deterministic dynamics of distributional multi-agent reinforcement learning
PLOS Computational Biology · 2026
DOI: 10.1371/journal.pcbi.1014723 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2026. Verfasst von Clémence Bergerot und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biology, Humboldt Universität zu Berlin, Berlin, Germany. Hauptthemen laut MEDLINE: Reinforcement Machine Learning, Reinforcement, Psychology. Die Arbeit nennt 95 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Einstein Center for Neurosciences Berlin, Deutsche Forschungsgemeinschaft, Cooperative AI Foundation.
Abstract (Originalsprache)
Understanding how cognition shapes behavior across contexts remains a fundamental challenge for many disciplines. In particular, for the optimism heuristic–i.e., the tendency to overweight positive (relative to negative) information–knowledge remains fragmented, with models developed in specific domains in isolation. Here, we present a unifying computational framework by deriving the deterministic dynamics of distributional multi-agent reinforcement learning. Our approach discretizes return distributions through a finite set of neurons, consistent with recent empirical findings on distributional coding in the brain. We validate our framework by reproducing established results across three iconic domains spanning individual bandit choice under resource variability, social coordination, and risky choice. Beyond validation, we uncover novel interactions among optimism, return discretization, and temporal discounting. Specifically, we identify conditions under which return discretization generates choice hysteresis and, in extreme parameter regimes, inescapable perseveration. We further reveal “individual dilemmas”: circumstances where agents gravitate toward suboptimal yet stable strategies, offering a mechanistic explanation for incoherent choice patterns. Our framework bridges neuroscience, psychology, and collective behavior, enabling empirically testable hypotheses about how cognitive biases propagate from individual cognition to social outcomes in complex environments.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Biology, Humboldt Universität zu Berlin, Berlin, Germany.
- Charité-Universitätsmedizin Berlin, Einstein Center for Neurosciences Berlin, Berlin, Germany.
- Science of Intelligence, Research Cluster of Excellence, Berlin, Germany.
- Transdisciplinary Research Area Sustainable Futures, University of Bonn, Bonn, Germany.
- Center for Development Research, University of Bonn, Bonn, Germany.
- Institute for Food and Resource Economics, University of Bonn, Bonn, Germany.
ORCID-Profile: Clémence Bergerot, Wolfram Barfuss
Förderung
- Einstein Center for Neurosciences Berlin
- Deutsche Forschungsgemeinschaft
- Cooperative AI Foundation
Zitieren
APA
Bergerot, C., Romanczuk, P., & Barfuss, W. (2026). Deterministic dynamics of distributional multi-agent reinforcement learning. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1014723
Vancouver
Bergerot C, Romanczuk P, Barfuss W. Deterministic dynamics of distributional multi-agent reinforcement learning. PLOS Computational Biology. 2026. doi:10.1371/journal.pcbi.1014723
BibTeX
@article{Bergerot2026,
author = {Bergerot, Clémence and Romanczuk, Pawel and Barfuss, Wolfram},
title = {Deterministic dynamics of distributional multi-agent reinforcement learning},
journal = {PLOS Computational Biology},
year = {2026},
doi = {10.1371/journal.pcbi.1014723}
}
EUVIMED