Randomized Exploration in Cooperative Multi-Agent Reinforcement Learning

arXiv:2404.10728v1 Announce Type: new
Abstract: We present the first study on provably efficient randomized exploration in cooperative multi-agent reinforcement learning (MARL). We propose a unified algorithm framework for randomized exploration in parallel Markov Decision Processes (MDPs), and two Thompson Sampling (TS)-type algorithms, CoopTS-PHE and CoopTS-LMC, incorporating the perturbed-history exploration (PHE) strategy and the Langevin Monte Carlo exploration (LMC) strategy respectively, which are flexible in design and easy to implement in practice. For a special class of parallel MDPs where the …

abstract agent algorithm algorithms arxiv cs.lg decision exploration framework history markov multi-agent processes reinforcement reinforcement learning sampling stat.ml strategy study type