Mitigating Relative Over-Generalization in Multi-Agent Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Ting, Jin, Yue, Houssineau, Jeremie, Montana, Giovanni
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910702463090688
author Zhu, Ting
Jin, Yue
Houssineau, Jeremie
Montana, Giovanni
author_facet Zhu, Ting
Jin, Yue
Houssineau, Jeremie
Montana, Giovanni
contents In decentralized multi-agent reinforcement learning, agents learning in isolation can lead to relative over-generalization (RO), where optimal joint actions are undervalued in favor of suboptimal ones. This hinders effective coordination in cooperative tasks, as agents tend to choose actions that are individually rational but collectively suboptimal. To address this issue, we introduce MaxMax Q-Learning (MMQ), which employs an iterative process of sampling and evaluating potential next states, selecting those with maximal Q-values for learning. This approach refines approximations of ideal state transitions, aligning more closely with the optimal joint policy of collaborating agents. We provide theoretical analysis supporting MMQ's potential and present empirical evaluations across various environments susceptible to RO. Our results demonstrate that MMQ frequently outperforms existing baselines, exhibiting enhanced convergence and sample efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2411_11099
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Mitigating Relative Over-Generalization in Multi-Agent Reinforcement Learning
Zhu, Ting
Jin, Yue
Houssineau, Jeremie
Montana, Giovanni
Machine Learning
Artificial Intelligence
In decentralized multi-agent reinforcement learning, agents learning in isolation can lead to relative over-generalization (RO), where optimal joint actions are undervalued in favor of suboptimal ones. This hinders effective coordination in cooperative tasks, as agents tend to choose actions that are individually rational but collectively suboptimal. To address this issue, we introduce MaxMax Q-Learning (MMQ), which employs an iterative process of sampling and evaluating potential next states, selecting those with maximal Q-values for learning. This approach refines approximations of ideal state transitions, aligning more closely with the optimal joint policy of collaborating agents. We provide theoretical analysis supporting MMQ's potential and present empirical evaluations across various environments susceptible to RO. Our results demonstrate that MMQ frequently outperforms existing baselines, exhibiting enhanced convergence and sample efficiency.
title Mitigating Relative Over-Generalization in Multi-Agent Reinforcement Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2411.11099