Represented Value Function Approach for Large Scale Multi Agent Reinforcement Learning
This addresses scalability issues in multi-agent systems for domains like gaming, but it appears incremental as it builds on existing value function methods.
The paper tackles large-scale multi-agent reinforcement learning by representing pairwise value functions to reduce interaction complexity and using an l2-norm trick to bound approximations, with experiments on a battle game showing effectiveness.
In this paper, we consider the problem of large scale multi agent reinforcement learning. Firstly, we studied the representation problem of the pairwise value function to reduce the complexity of the interactions among agents. Secondly, we adopt a l2-norm trick to ensure the trivial term of the approximated value function is bounded. Thirdly, experimental results on battle game demonstrate the effectiveness of the proposed approach.