Provably Efficient Reinforcement Learning for Adversarial Restless Multi-Armed Bandits with Unknown Transitions and Bandit Feedback.

Guojun Xiong, Jian Li

Browse the full ICML paper archive.