Learning Adversarial MDPs with Bandit Feedback and Unknown Transition
| Autor: | Jin, Chi, Jin, Tiancheng, Luo, Haipeng, Sra, Suvrit, Yu, Tiancheng |
|---|---|
| Rok vydání: | 2019 |
| Předmět: | |
| Druh dokumentu: | Working Paper |
| Popis: | We consider the problem of learning in episodic finite-horizon Markov decision processes with an unknown transition function, bandit feedback, and adversarial losses. We propose an efficient algorithm that achieves $\mathcal{\tilde{O}}(L|X|\sqrt{|A|T})$ regret with high probability, where $L$ is the horizon, $|X|$ is the number of states, $|A|$ is the number of actions, and $T$ is the number of episodes. To the best of our knowledge, our algorithm is the first to ensure $\mathcal{\tilde{O}}(\sqrt{T})$ regret in this challenging setting; in fact it achieves the same regret bound as (Rosenberg & Mansour, 2019a) that considers an easier setting with full-information feedback. Our key technical contributions are two-fold: a tighter confidence set for the transition function, and an optimistic loss estimator that is inversely weighted by an $\textit{upper occupancy bound}$. Comment: Fix a bug |
| Databáze: | arXiv |
| Externí odkaz: |
|