Greedy-based Value Representation for Optimal Coordination in Multi-agent Reinforcement Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wan, Lipeng, Liu, Zeyang, Chen, Xingyu, Wang, Han, Lan, Xuguang
Format: Preprint
Veröffentlicht: 2021
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910040505450496
author Wan, Lipeng
Liu, Zeyang
Chen, Xingyu
Wang, Han
Lan, Xuguang
author_facet Wan, Lipeng
Liu, Zeyang
Chen, Xingyu
Wang, Han
Lan, Xuguang
contents Due to the representation limitation of the joint Q value function, multi-agent reinforcement learning methods with linear value decomposition (LVD) or monotonic value decomposition (MVD) suffer from relative overgeneralization. As a result, they can not ensure optimal consistency (i.e., the correspondence between individual greedy actions and the maximal true Q value). In this paper, we derive the expression of the joint Q value function of LVD and MVD. According to the expression, we draw a transition diagram, where each self-transition node (STN) is a possible convergence. To ensure optimal consistency, the optimal node is required to be the unique STN. Therefore, we propose the greedy-based value representation (GVR), which turns the optimal node into an STN via inferior target shaping and further eliminates the non-optimal STNs via superior experience replay. In addition, GVR achieves an adaptive trade-off between optimality and stability. Our method outperforms state-of-the-art baselines in experiments on various benchmarks. Theoretical proofs and empirical results on matrix games demonstrate that GVR ensures optimal consistency under sufficient exploration.
format Preprint
id arxiv_https___arxiv_org_abs_2112_04454
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Greedy-based Value Representation for Optimal Coordination in Multi-agent Reinforcement Learning
Wan, Lipeng
Liu, Zeyang
Chen, Xingyu
Wang, Han
Lan, Xuguang
Multiagent Systems
Due to the representation limitation of the joint Q value function, multi-agent reinforcement learning methods with linear value decomposition (LVD) or monotonic value decomposition (MVD) suffer from relative overgeneralization. As a result, they can not ensure optimal consistency (i.e., the correspondence between individual greedy actions and the maximal true Q value). In this paper, we derive the expression of the joint Q value function of LVD and MVD. According to the expression, we draw a transition diagram, where each self-transition node (STN) is a possible convergence. To ensure optimal consistency, the optimal node is required to be the unique STN. Therefore, we propose the greedy-based value representation (GVR), which turns the optimal node into an STN via inferior target shaping and further eliminates the non-optimal STNs via superior experience replay. In addition, GVR achieves an adaptive trade-off between optimality and stability. Our method outperforms state-of-the-art baselines in experiments on various benchmarks. Theoretical proofs and empirical results on matrix games demonstrate that GVR ensures optimal consistency under sufficient exploration.
title Greedy-based Value Representation for Optimal Coordination in Multi-agent Reinforcement Learning
topic Multiagent Systems
url https://arxiv.org/abs/2112.04454