Multi-Agent Actor-Critic with Hierarchical Graph Attention Network
This addresses the challenge of applying trained policies to more complex multi-agent tasks, though it appears incremental as it builds on existing multi-agent reinforcement learning methods.
The paper tackles the problem of limited transferability in multi-agent reinforcement learning policies to new tasks by proposing a model combining hierarchical graph attention networks with multi-agent actor-critic, which outperforms existing methods in mixed cooperative and competitive tasks.
Most previous studies on multi-agent reinforcement learning focus on deriving decentralized and cooperative policies to maximize a common reward and rarely consider the transferability of trained policies to new tasks. This prevents such policies from being applied to more complex multi-agent tasks. To resolve these limitations, we propose a model that conducts both representation learning for multiple agents using hierarchical graph attention network and policy learning using multi-agent actor-critic. The hierarchical graph attention network is specially designed to model the hierarchical relationships among multiple agents that either cooperate or compete with each other to derive more advanced strategic policies. Two attention networks, the inter-agent and inter-group attention layers, are used to effectively model individual and group level interactions, respectively. The two attention networks have been proven to facilitate the transfer of learned policies to new tasks with different agent compositions and allow one to interpret the learned strategies. Empirically, we demonstrate that the proposed model outperforms existing methods in several mixed cooperative and competitive tasks.