Teaching on a Budget in Multi-Agent Deep Reinforcement Learning
This work addresses sample efficiency for researchers in cooperative decentralized multi-agent reinforcement learning, but it is incremental as it builds on existing teacher-student frameworks with limited novelty.
The paper tackles the problem of poor sample efficiency in multi-agent deep reinforcement learning by proposing heuristics-based action advising techniques within a teacher-student framework, showing experimental results in a gridworld environment that suggest the approach is useful and warrants further investigation.
Deep Reinforcement Learning (RL) algorithms can solve complex sequential decision tasks successfully. However, they have a major drawback of having poor sample efficiency which can often be tackled by knowledge reuse. In Multi-Agent Reinforcement Learning (MARL) this drawback becomes worse, but at the same time, a new set of opportunities to leverage knowledge are also presented through agent interactions. One promising approach among these is peer-to-peer action advising through a teacher-student framework. Despite being introduced for single-agent RL originally, recent studies show that it can also be applied to multi-agent scenarios with promising empirical results. However, studies in this line of research are currently very limited. In this paper, we propose heuristics-based action advising techniques in cooperative decentralised MARL, using a nonlinear function approximation based task-level policy. By adopting Random Network Distillation technique, we devise a measurement for agents to assess their knowledge in any given state and be able to initiate the teacher-student dynamics with no prior role assumptions. Experimental results in a gridworld environment show that such an approach may indeed be useful and needs to be further investigated.