On Feasible Rewards in Multi-Agent Inverse Reinforcement Learning

Till Freihaut, Giorgia Ramponi

Keywords: Reinforcement Learning, Multi-Agent Reinforcement Learning, Inverse Reinforcement Learning, Multi-Agent Inverse Reinforcement Learning

Abstract: Multi-agent inverse reinforcement learning (MAIRL) aims to recover agent reward functions from expert demonstrations. We characterize the feasible reward set in Markov games, identifying all reward functions that rationalize a given equilibrium. However, equilibrium-based observations are often ambiguous: a single Nash equilibrium can correspond to many reward structures, potentially changing the game's nature in multi-agent systems. We address this by introducing entropy-regularized Markov games, which yield a unique equilibrium while preserving strategic incentives. For this setting, we provide a sample complexity analysis detailing how errors affect learned policy performance. Our work establishes theoretical foundations and practical insights for MAIRL.

Primary Area: Reinforcement learning (e.g., decision and control, planning, hierarchical RL, robotics)

Submission Number: 13429