TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play
It addresses the unsolved problem of multi-agent coordination and planning in complex environments for AI research, with incremental innovations applied to football and other domains.
The paper tackled the challenge of multi-agent football by developing TiZero, a system that learns to play the full 11 vs. 11 game without demonstrations, achieving over 30% higher win rates than previous systems.
Multi-agent football poses an unsolved challenge in AI research. Existing work has focused on tackling simplified scenarios of the game, or else leveraging expert demonstrations. In this paper, we develop a multi-agent system to play the full 11 vs. 11 game mode, without demonstrations. This game mode contains aspects that present major challenges to modern reinforcement learning algorithms; multi-agent coordination, long-term planning, and non-transitivity. To address these challenges, we present TiZero; a self-evolving, multi-agent system that learns from scratch. TiZero introduces several innovations, including adaptive curriculum learning, a novel self-play strategy, and an objective that optimizes the policies of multiple agents jointly. Experimentally, it outperforms previous systems by a large margin on the Google Research Football environment, increasing win rates by over 30%. To demonstrate the generality of TiZero's innovations, they are assessed on several environments beyond football; Overcooked, Multi-agent Particle-Environment, Tic-Tac-Toe and Connect-Four.