Settling Decentralized Multi-Agent Coordinated Exploration by Novelty Sharing
This addresses exploration challenges for decentralized multi-agent systems, offering a novel approach to improve coordination, though it is incremental in nature.
The paper tackles the problem of coordinated exploration in decentralized multi-agent reinforcement learning by proposing MACE, a method that shares local novelty and uses weighted mutual information as an intrinsic reward, achieving superior performance in three sparse-reward environments.
Exploration in decentralized cooperative multi-agent reinforcement learning faces two challenges. One is that the novelty of global states is unavailable, while the novelty of local observations is biased. The other is how agents can explore in a coordinated way. To address these challenges, we propose MACE, a simple yet effective multi-agent coordinated exploration method. By communicating only local novelty, agents can take into account other agents' local novelty to approximate the global novelty. Further, we newly introduce weighted mutual information to measure the influence of one agent's action on other agents' accumulated novelty. We convert it as an intrinsic reward in hindsight to encourage agents to exert more influence on other agents' exploration and boost coordinated exploration. Empirically, we show that MACE achieves superior performance in three multi-agent environments with sparse rewards.