Herd’s eye view: improving game AI agent learning with collaborative perception

Abstract

We present a novel perception model named Herd’s Eye View that adopts a global perspective derived from multiple agents to boost the decision-making capabilities of reinforcement learning agents in multi-agent environments, specifically in the context of game AI. The Herd’s Eye View approach utilizes cooperative perception to empower reinforcement learning agents with global reasoning ability, enhancing their decision- making. We demonstrate the effectiveness of Herd’s Eye View within simulated game environments and highlight its superior performance compared to traditional egocentric perception models. This work contributes to cooperative perception and multi- agent reinforcement learning by offering a more realistic and efficient perspective for global coordination and decision-making within game environments. Moreover, our approach promotes broader AI applications beyond gaming by addressing constraints faced by AI in other fields such as robotics.