Inverse Reinforcement Learning for Multiplayer Non-Zero-Sum Games
摘要
ChaptersReinforcement Learning (RL)chapter6 andchapter7 focus on inverse RL for differential systems involving one or two players, where each player’s input represents their decision-making outcome. There is also a class of real-world dynamic applications that involve multiple control . For instance, in vehicle operation, various controls such as steering wheel, pedals for acceleration and braking, turn signals, and gear selectors are involved. This motivates us to explore the scenario of multiple control players in a single dynamic system.