Approximate Linear Programming for Decentralized Policy Iteration in Cooperative Multi-agent Markov Decision Processes

arXiv:2311.11789v2 Announce Type: replace
Abstract: In this work, we consider a cooperative multi-agent Markov decision process (MDP) involving m agents. At each decision epoch, all the m agents independently select actions in order to maximize a common long-term objective. In the policy iteration process of multi-agent setup, the number of actions grows exponentially with the number of agents, incurring huge computational costs. Thus, recent works consider decentralized policy improvement, where each agent improves its decisions unilaterally, assuming that the decisions …

abstract agent agents arxiv cs.lg cs.ma decentralized decision iteration linear long-term markov math.oc multi-agent policy process processes programming setup type work