A further exploration of deep Multi-Agent Reinforcement Learning with Hybrid Action Space

TL;DR

This paper introduces two novel algorithms, MAHSAC and MAHDDPG, designed to enable multi-agent deep reinforcement learning in environments with hybrid action spaces, addressing a gap in real-world applications.

Contribution

The paper proposes two algorithms that extend multi-agent deep reinforcement learning to hybrid action spaces, a previously underexplored area.

Findings

Algorithms perform well in multi-agent particle environment

Effective handling of hybrid action spaces demonstrated

Contributes to real-world multi-agent RL applications

Abstract

The research of extending deep reinforcement learning (drl) to multi-agent field has solved many complicated problems and made great achievements. However, almost all these studies only focus on discrete or continuous action space and there are few works having ever used multi-agent deep reinforcement learning to real-world environment problems which mostly have a hybrid action space. Therefore, in this paper, we propose two algorithms: deep multi-agent hybrid soft actor-critic (MAHSAC) and multi-agent hybrid deep deterministic policy gradients (MAHDDPG) to fill this gap. This two algorithms follow the centralized training and decentralized execution (CTDE) paradigm and could handle hybrid action space problems. Our experiences are running on multi-agent particle environment which is an easy multi-agent particle world, along with some basic simulated physics. The experimental results show that these algorithms have good performances.