Learning to Share and Hide Intentions using Information Regularization

TL;DR

This paper introduces an information regularization method enabling agents to learn cooperative or competitive strategies in asymmetric information games without direct interaction or models of other agents.

Contribution

It proposes a novel approach to control information sharing in multi-agent reinforcement learning using information-theoretic regularizers, applicable without agent interaction.

Findings

Agents can learn effective cooperation and competition strategies.

Regularizers improve reward outcomes for the second agent.

Method integrates easily with policy gradient reinforcement learning.

Abstract

Learning to cooperate with friends and compete with foes is a key component of multi-agent reinforcement learning. Typically to do so, one requires access to either a model of or interaction with the other agent(s). Here we show how to learn effective strategies for cooperation and competition in an asymmetric information game with no such model or interaction. Our approach is to encourage an agent to reveal or hide their intentions using an information-theoretic regularizer. We consider both the mutual information between goal and action given state, as well as the mutual information between goal and state. We show how to optimize these regularizers in a way that is easy to integrate with policy gradient reinforcement learning. Finally, we demonstrate that cooperative (competitive) policies learned with our approach lead to more (less) reward for a second agent in two simple asymmetric information games.