A Comprehensive Review of Multiagent Reinforcement Learning in Video Games

Citation: Z. Li, Q. Ji, X. Ling and Q. Liu, "A Comprehensive Review of Multiagent Reinforcement Learning in Video Games," in IEEE Transactions on Games, vol. 17, no. 4, pp. 873-892, Dec. 2025, doi: 10.1109/TG.2025.3588809.

Recent advancements in multiagent reinforcement learning (MARL) have demonstrated its application potential in modern games. Beginning with foundational work and progressing to landmark achievements, such as AlphaStar in StarCraft II and OpenAI Five in Dota 2, MARL has proven capable of achieving superhuman performance across diverse game environments through techniques, such as self-play, supervised learning, and deep reinforcement learning. With its growing impact, a comprehensive review has become increasingly important in this field. This article aims to provide a thorough examination of MARL’s application from turn-based two-agent games to real-time multiagent video games, including popular genres, such as sports games, first-person shooter games, real-time strategy games and multiplayer online battle arena games. We further analyze critical challenges posed by MARL in video games, including nonstationary, partial observability, sparse rewards, team coordination, and scalability, and highlight successful implementations in games, such as Rocket League, Minecraft, Quake III Arena, StarCraft II, Dota 2, Honor of Kings, etc. This article offers insights into MARL in video game AI systems, proposes a novel method to estimate game complexity, and suggests future research directions to advance MARL and its applications in game development, inspiring further innovation in this rapidly evolving field.

Read about it here