Gahee Kim

MA$^2$E: Addressing Partial Observability in Multi-Agent Reinforcement Learning with Masked Auto-Encoder

Centralized Training and Decentralized Execution (CTDE) is a widely adopted paradigm to solve cooperative multi-agent reinforcement learning (MARL) problems. Despite the successes achieved with CTDE, partial observability still limits cooperation among agents. While previous studies have attempted t…