Conditionally Optimistic Exploration for Cooperative Deep Multi-Agent Reinforcement Learning

TL;DR

This paper introduces Conditionally Optimistic Exploration (COE), a novel exploration method for cooperative multi-agent reinforcement learning that leverages sequential action computation and tree search principles to improve exploration efficiency.

Contribution

The paper proposes COE, a new exploration approach inspired by UCT, which enhances cooperative exploration in MARL by incorporating optimistic bonuses based on sequential agent actions.

Findings

COE outperforms existing exploration methods on challenging MARL benchmarks.

COE is compatible with various value decomposition methods for centralized training.

The method effectively captures structured dependencies among agents during exploration.

Abstract

Efficient exploration is critical in cooperative deep Multi-Agent Reinforcement Learning (MARL). In this work, we propose an exploration method that effectively encourages cooperative exploration based on the idea of sequential action-computation scheme. The high-level intuition is that to perform optimism-based exploration, agents would explore cooperative strategies if each agent's optimism estimate captures a structured dependency relationship with other agents. Assuming agents compute actions following a sequential order at \textit{each environment timestep}, we provide a perspective to view MARL as tree search iterations by considering agents as nodes at different depths of the search tree. Inspired by the theoretically justified tree search algorithm UCT (Upper Confidence bounds applied to Trees), we develop a method called Conditionally Optimistic Exploration (COE). COE augments each agent's state-action value estimate with an action-conditioned optimistic bonus derived from the visitation count of the global state and joint actions of preceding agents. COE is performed during training and disabled at deployment, making it compatible with any value decomposition method for centralized training with decentralized execution. Experiments across various cooperative MARL benchmarks show that COE outperforms current state-of-the-art exploration methods on hard-exploration tasks.