Value-Decomposition Networks For Cooperative Multi-Agent Learning
This addresses challenges in multi-agent systems for applications like robotics or gaming, though it is incremental as it builds on existing value-based methods.
The paper tackles the problem of cooperative multi-agent reinforcement learning with a single joint reward signal by addressing issues like spurious rewards and the 'lazy agent' problem through a value decomposition network architecture that learns to decompose team value into agent-wise functions, showing superior results in partially-observable domains.
We study the problem of cooperative multi-agent reinforcement learning with a single joint reward signal. This class of learning problems is difficult because of the often large combined action and observation spaces. In the fully centralized and decentralized approaches, we find the problem of spurious rewards and a phenomenon we call the "lazy agent" problem, which arises due to partial observability. We address these problems by training individual agents with a novel value decomposition network architecture, which learns to decompose the team value function into agent-wise value functions. We perform an experimental evaluation across a range of partially-observable multi-agent domains and show that learning such value-decompositions leads to superior results, in particular when combined with weight sharing, role information and information channels.