Projects per year
Abstract
Relative overgeneralization (RO) occurs in cooperative multi-agent learning tasks when agents converge towards a suboptimal joint policy due to overfitting to suboptimal behaviors of other agents.No methods have been proposed for addressing RO in multi-agent policy gradient (MAPG) methods although these methods produce state-of-the-art results.To address this gap, we propose a general, yet simple, framework to enable optimistic updates in MAPG methods that alleviate the RO problem.Our approach involves clipping the advantage to eliminate negative values, thereby facilitating optimistic updates in MAPG.The optimism prevents individual agents from quickly converging to a local optimum.Additionally, we provide a formal analysis to show that the proposed method retains optimality at a fixed point.In extensive evaluations on a diverse set of tasks including the Multi-agent MuJoCo and Overcooked benchmarks, our method outperforms strong baselines on 13 out of 19 tested tasks and matches the performance on the rest.
Original language | English |
---|---|
Pages (from-to) | 61186-61202 |
Number of pages | 17 |
Journal | Proceedings of Machine Learning Research |
Volume | 235 |
Publication status | Published - 2024 |
MoE publication type | A4 Conference publication |
Event | International Conference on Machine Learning - Vienna, Austria Duration: 21 Jul 2024 → 27 Jul 2024 Conference number: 41 |
Fingerprint
Dive into the research topics of 'Optimistic Multi-Agent Policy Gradient'. Together they form a unique fingerprint.-
MARL: Efficient and Principled Multi-Agent Reinforcement Learning
Pajarinen, J. (Principal investigator), Yang, W. (Project Member), Zhao, W. (Project Member) & Tuomisto, J. (Project Member)
01/09/2023 → 31/08/2027
Project: Academy of Finland: Other research funding
-
Energy efficient machine perception /Kannala 31.12.2025: Energy efficient perception
Kannala, J. (Principal investigator), Turkulainen, M. (Project Member), Fang, J. (Project Member), Zhang, Y. (Project Member) & Ferranti, L. (Project Member)
01/01/2023 → 31/12/2025
Project: Academy of Finland: Other research funding
-
REPEAT: Robust and Efficient PErception for Autonomous Things
Kannala, J. (Principal investigator), Ye, R. (Project Member), Boney, R. (Project Member), Li, X. (Project Member), Melekhov, I. (Project Member), Fang, J. (Project Member), Zhang, Y. (Project Member) & Krahn, M. (Project Member)
01/01/2020 → 30/09/2023
Project: Academy of Finland: Other research funding