Effects of Spectral Normalization in Multi-agent Reinforcement Learning

Autor:	Mehta, Kinal, Mahajan, Anuj, Kumar, Pawan
Rok vydání:	2022
Předmět:	Computer Science - Machine Learning Computer Science - Artificial Intelligence
Druh dokumentu:	Working Paper
Popis:	A reliable critic is central to on-policy actor-critic learning. But it becomes challenging to learn a reliable critic in a multi-agent sparse reward scenario due to two factors: 1) The joint action space grows exponentially with the number of agents 2) This, combined with the reward sparseness and environment noise, leads to large sample requirements for accurate learning. We show that regularising the critic with spectral normalization (SN) enables it to learn more robustly, even in multi-agent on-policy sparse reward scenarios. Our experiments show that the regularised critic is quickly able to learn from the sparse rewarding experience in the complex SMAC and RWARE domains. These findings highlight the importance of regularisation in the critic for stable learning. Comment: Accepted at IJCNN-2023
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2212.05331 Zobrazit plný text záznamu View this record from Arxiv