Entropy Seeking Constrained Multiagent Reinforcement Learning

Ayhan Alp Aydeniz (Collaborative Robotics and Intelligent Systems Institute, Oregon State University), Enrico Marchesini (Laboratory for Information Decision Systems, Massachusetts Institute of Technology), Christopher Amato (Khoury College of Computer Sciences, Northeastern University), Kagan Tumer (Collaborative Robotics and Intelligent Systems Institute, Oregon State University)

Abstract

Multiagent Reinforcement Learning (MARL) has been successfully applied to domains requiring close coordination among many agents. However, real-world tasks require safety specifications that are not generally considered by MARL algorithms. In this work, we introduce an Entropy Seeking Constrained (ESC) approach aiming to learn safe cooperative policies for multiagent systems. Unlike previous methods, ESC considers safety specifications while maximizing state-visitation entropy, addressing the exploration issues of constrained-based solutions.