Learning to Incentivize Other Learning Agents
This addresses the challenge of ensuring cooperation for the common good in multi-agent reinforcement learning systems, representing a novel approach rather than an incremental improvement.
The paper tackles the problem of cooperation among independent reinforcement learning agents in shared multi-agent environments by enabling agents to learn incentive functions that reward others, leading to significant performance improvements over standard RL and opponent-shaping agents in challenging general-sum Markov games, often achieving near-optimal division of labor.
The challenge of developing powerful and general Reinforcement Learning (RL) agents has received increasing attention in recent years. Much of this effort has focused on the single-agent setting, in which an agent maximizes a predefined extrinsic reward function. However, a long-term question inevitably arises: how will such independent agents cooperate when they are continually learning and acting in a shared multi-agent environment? Observing that humans often provide incentives to influence others' behavior, we propose to equip each RL agent in a multi-agent environment with the ability to give rewards directly to other agents, using a learned incentive function. Each agent learns its own incentive function by explicitly accounting for its impact on the learning of recipients and, through them, the impact on its own extrinsic objective. We demonstrate in experiments that such agents significantly outperform standard RL and opponent-shaping agents in challenging general-sum Markov games, often by finding a near-optimal division of labor. Our work points toward more opportunities and challenges along the path to ensure the common good in a multi-agent future.