Inverse reinforcement learning for multi-player noncooperative apprentice games

In this paper, we devise inverse reinforcement learning (RL) algorithms for nonlinear continuous-time systems described by multiplayer differential equations. We define a new class of Multi-player Noncooperative Apprentice Games, in which both the expert and learner have N-player control inputs. The games are solved learners reconstructing unknown performance reward functions experts from experts’ trajectories, i.e., states optimal first develop model-based RL algorithm that involves two stages: an stage second (IOC) stage. Our solves IOC as subproblem. therefore provide one possible unified framework dynamic systems. then using neural networks: completely model-free homogeneous inputs; partially heterogeneous Finally, present results simulations, verify validity our proposed algorithms.