VDFD: Multi-Agent Value Decomposition Framework with Disentangled World Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Zhizun, Meger, David
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909806078459904
author Wang, Zhizun
Meger, David
author_facet Wang, Zhizun
Meger, David
contents In this paper, we propose a novel model-based multi-agent reinforcement learning approach named Value Decomposition Framework with Disentangled World Model to address the challenge of achieving a common goal of multiple agents interacting in the same environment with reduced sample complexity. Due to scalability and non-stationarity problems posed by multi-agent systems, model-free methods rely on a considerable number of samples for training. In contrast, we use a modularized world model, composed of action-conditioned, action-free, and static branches, to unravel the complicated environment dynamics. Our model produces imagined outcomes based on past experience, without sampling directly from the real environment. We employ variational auto-encoders and variational graph auto-encoders to learn the latent representations for the world model, which is merged with a value-based framework to predict the joint action-value function and optimize the overall training objective. Experimental results on StarCraft II micro-management, Multi-Agent MuJoCo, and Level-Based Foraging challenges demonstrate that our method achieves high sample efficiency and exhibits superior performance compared to other baselines across a wide range of multi-agent learning tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2309_04615
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle VDFD: Multi-Agent Value Decomposition Framework with Disentangled World Model
Wang, Zhizun
Meger, David
Machine Learning
Artificial Intelligence
Multiagent Systems
In this paper, we propose a novel model-based multi-agent reinforcement learning approach named Value Decomposition Framework with Disentangled World Model to address the challenge of achieving a common goal of multiple agents interacting in the same environment with reduced sample complexity. Due to scalability and non-stationarity problems posed by multi-agent systems, model-free methods rely on a considerable number of samples for training. In contrast, we use a modularized world model, composed of action-conditioned, action-free, and static branches, to unravel the complicated environment dynamics. Our model produces imagined outcomes based on past experience, without sampling directly from the real environment. We employ variational auto-encoders and variational graph auto-encoders to learn the latent representations for the world model, which is merged with a value-based framework to predict the joint action-value function and optimize the overall training objective. Experimental results on StarCraft II micro-management, Multi-Agent MuJoCo, and Level-Based Foraging challenges demonstrate that our method achieves high sample efficiency and exhibits superior performance compared to other baselines across a wide range of multi-agent learning tasks.
title VDFD: Multi-Agent Value Decomposition Framework with Disentangled World Model
topic Machine Learning
Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2309.04615