Unilateral incentive alignment in two-agent stochastic games
Proceedings of the National Academy of Sciences · 24.06.2025 · 122 (25)
DOI: 10.1073/pnas.2319927121 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 16.06.2025. Verfasst von Alex McAvoy und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Data Science and Society, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Multiagent learning is challenging when agents face mixed-motivation interactions, where conflicts of interest arise as agents independently try to optimize their respective outcomes. Recent advancements in evolutionary game theory have identified a class of “zero-determinant” strategies, which confer an agent with significant unilateral control over outcomes in repeated games. Building on these insights, we present a comprehensive generalization of zero-determinant strategies to stochastic games, encompassing dynamic environments. We propose an algorithm that allows an agent to discover strategies enforcing predetermined linear (or approximately linear) payoff relationships. Of particular interest is the relationship in which both payoffs are equal, which serves as a proxy for fairness in symmetric games. We demonstrate that an agent can discover strategies enforcing such relationships through experience alone, without coordinating with an opponent. In finding and using such a strategy, an agent (“enforcer”) can incentivize optimal and equitable outcomes, circumventing potential exploitation. In particular, from the opponent’s viewpoint, the enforcer transforms a mixed-motivation problem into a cooperative problem, paving the way for more collaboration and fairness in multiagent systems.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Data Science and Society, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599.
- Department of Mathematics, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599.
- Department of Computer Science, Stanford University, Stanford, CA 94305.
- JP Morgan AI Research, New York, NY 10017.
- Max Planck Research Group: Dynamics of Social Behavior, Max Planck Institute for Evolutionary Biology, Plön 24306, Germany.
- Institute of Science and Technology Austria, Klosterneuburg 3400, Austria.
ORCID-Profile: Alex McAvoy, Udari Madhushani Sehwag, Christian Hilbe, Krishnendu Chatterjee, Wolfram Barfuss, Qi Su, Naomi Ehrich Leonard, Joshua B Plotkin
Zitieren
APA
McAvoy, A., Madhushani Sehwag, U., Hilbe, C., Chatterjee, K., Barfuss, W., Su, Q., Leonard, N. E., & Plotkin, J. B. (2025). Unilateral incentive alignment in two-agent stochastic games. Proceedings of the National Academy of Sciences, 122(25). https://doi.org/10.1073/pnas.2319927121
Vancouver
McAvoy A, Madhushani Sehwag U, Hilbe C, Chatterjee K, Barfuss W, Su Q, et al. Unilateral incentive alignment in two-agent stochastic games. Proceedings of the National Academy of Sciences. 2025;122(25). doi:10.1073/pnas.2319927121
BibTeX
@article{McAvoy2025,
author = {McAvoy, Alex and Madhushani Sehwag, Udari and Hilbe, Christian and Chatterjee, Krishnendu and Barfuss, Wolfram and Su, Qi and Leonard, Naomi Ehrich and Plotkin, Joshua B},
title = {Unilateral incentive alignment in two-agent stochastic games},
journal = {Proceedings of the National Academy of Sciences},
year = {2025},
volume = {122},
number = {25},
doi = {10.1073/pnas.2319927121}
}
EUVIMED