Revisit minimax lower bounds in episodic rl in finite mdps
Hao Liang (CUHK-SZ)
Speaker: Hao Liang
Title: Revisit minimax lower bounds in episodic rl in finite mdps
Time: Jul 9 2pm-5pm
Reference:
Domingues, O. D., Ménard, P., Kaufmann, E., & Valko, M. (2021, March). Episodic reinforcement learning in finite mdps: Minimax lower bounds revisited. In Algorithmic Learning Theory (pp. 578-598). PMLR.