Regret Bounds for Risk-Sensitive Reinforcement Learning

Autor:	Bastani, O., Ma, Y. J., Shen, E., Xu, W.
Rok vydání:	2022
Předmět:	Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	In safety-critical applications of reinforcement learning such as healthcare and robotics, it is often desirable to optimize risk-sensitive objectives that account for tail outcomes rather than expected reward. We prove the first regret bounds for reinforcement learning under a general class of risk-sensitive objectives including the popular CVaR objective. Our theory is based on a novel characterization of the CVaR objective as well as a novel optimistic MDP construction.
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2210.05650 Zobrazit plný text záznamu View this record from Arxiv