Koopman Q-learning: Offline Reinforcement Learning via Symmetries of Dynamics.

Matthias Weissenbacher, Samarth Sinha, Animesh Garg, Yoshinobu Kawahara

Browse the full ICML paper archive.