Passer à la navigation principale Passer à la recherche Passer au contenu principal

Prediction-aware Learning in Multi-agent Systems

  • Ecole polytechnique
  • INRIA
  • PSL research University & IPSL

Résultats de recherche: Contribution à un journal › Article de conférence › Revue par des pairs

Résumé

The framework of uncoupled online learning in multiplayer games has made significant progress in recent years. In particular, the development of time-varying games has considerably expanded its modeling capabilities. However, current regret bounds quickly become vacuous when the game undergoes significant variations over time, even when these variations are easy to predict. Intuitively, the ability of players to forecast future payoffs should lead to tighter guarantees, yet existing approaches fail to incorporate this aspect. This work aims to fill this gap by introducing a novel prediction-aware framework for time-varying games, where agents can forecast future payoffs and adapt their strategies accordingly. In this framework, payoffs depend on an underlying state of nature that agents predict in an online manner. To leverage these predictions, we propose the POMWU algorithm, a contextual extension of the optimistic Multiplicative Weight Update algorithm, for which we establish theoretical guarantees on social welfare and convergence to equilibrium. Our results demonstrate that, under bounded prediction errors, the proposed framework achieves performance comparable to the static setting. Finally, we empirically demonstrate the effectiveness of POMWU in a traffic routing experiment.

langue originaleAnglais
Pages (de - à)6699-6725
Nombre de pages27
journalProceedings of Machine Learning Research
Volume267
étatPublié - 1 janv. 2025
Evénement42nd International Conference on Machine Learning, ICML 2025 - Vancouver, Canada
Durée: 13 juil. 2025 → 19 juil. 2025

Empreinte digitale

Examiner les sujets de recherche de « Prediction-aware Learning in Multi-agent Systems ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation