Multi-Agent Transfer Learning via Temporal Contrastive Learning
It addresses sample efficiency and interpretability in complex multi-agent coordination tasks, representing an incremental advancement in transfer learning methods.
This paper tackles the problem of transfer learning in deep multi-agent reinforcement learning by introducing a framework that combines goal-conditioned policies with temporal contrastive learning to discover sub-goals, resulting in improved sample efficiency and performance on tasks like Overcooked, requiring only 21.7% of training samples compared to baselines.
This paper introduces a novel transfer learning framework for deep multi-agent reinforcement learning. The approach automatically combines goal-conditioned policies with temporal contrastive learning to discover meaningful sub-goals. The approach involves pre-training a goal-conditioned agent, finetuning it on the target domain, and using contrastive learning to construct a planning graph that guides the agent via sub-goals. Experiments on multi-agent coordination Overcooked tasks demonstrate improved sample efficiency, the ability to solve sparse-reward and long-horizon problems, and enhanced interpretability compared to baselines. The results highlight the effectiveness of integrating goal-conditioned policies with unsupervised temporal abstraction learning for complex multi-agent transfer learning. Compared to state-of-the-art baselines, our method achieves the same or better performances while requiring only 21.7% of the training samples.