Graph Convolutional Reinforcement Learning

Jiechuan Jiang, Chen Dun, Tiejun Huang, Zongqing Lu
10/22/2018
cs.LGcs.AIcs.MA

Official code: PKU-RL/DGN (371 stars, last commit 2023-03-25).

Abstract

Learning to cooperate is crucially important in multi-agent environments. The key is to understand the mutual interplay between agents. However, multi-agent environments are highly dynamic, where agents keep moving and their neighbors change quickly. This makes it hard to learn abstract representations of mutual interplay between agents. To tackle these difficulties, we propose graph convolutional reinforcement learning, where graph convolution adapts to the dynamics of the underlying graph of the multi-agent environment, and relation kernels capture the interplay between agents by their relation representations. Latent features produced by convolutional layers from gradually increased receptive fields are exploited to learn cooperation, and cooperation is further improved by temporal relation regularization for consistency. Empirically, we show that our method substantially outperforms existing methods in a variety of cooperative scenarios.

View on arXivView PDF

Code Implementations(1)

PKU-RL/DGNOfficial100%

DGN Code

37187PythonOct 22, 20183 years ago

Similar Papers

Group-Aware Coordination Graph for Multi-Agent Reinforcement Learning

86% match1 repo322024

Inferring Latent Temporal Sparse Coordination Graph for Multi-Agent Reinforcement Learning

85% match1 repo72024

Learning Transferable Cooperative Behavior in Multi-Agent Teams

85% match1 repo1242019

TIGER-MARL: Enhancing Multi-Agent Reinforcement Learning with Temporal Information through Graph-based Embeddings and Representations

84% match2025

Deep Coordination Graphs

84% match1 repo772019

Enhancing Cooperation through Selective Interaction and Long-term Experiences in Multi-Agent Reinforcement Learning

84% match1 repo142024

Cite this paper

@article{jiang2018graph,
  title  = {Graph Convolutional Reinforcement Learning},
  author = {Jiechuan Jiang and Chen Dun and Tiejun Huang and Zongqing Lu},
  year   = {2018},
  eprint = {1810.09202},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1810.09202}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1810.09202.svg)](https://codeofpaper.com/paper/1810.09202)

Discussion