Recent Articles and Multimedia
Paper Title: Multi-agent Off-policy Actor-Critic Reinforcement Learning for Partially Observable Environments
Authors: Ainur Zhaikhan and Ali H. Sayed
When multiple agents are deployed across an environment, they may only have access to limited information, observing only specific portions of the overall environment. This study leverages a social learning method to estimate a global state within a multi-agent reinforcement learning framework operating in a partially observable environment.
READ MORE ABOUT THIS PAPER HERE
A block diagram illustrating the primary steps of the proposed algorithm.