Risk-Sensitive Reinforcement Learning: Near-Optimal Risk-Sample Tradeoff in Regret.
Yingjie FeiZhuoran YangYudong ChenZhaoran WangQiaomin XiePublished in: CoRR (2020)
Keyphrases
- risk sensitive
- reinforcement learning
- model free
- optimal control
- markov decision processes
- control policies
- reinforcement learning algorithms
- reward function
- markov decision problems
- risk neutral
- utility function
- function approximation
- optimal policy
- state space
- dynamic programming
- lower bound
- finite state
- temporal difference
- policy iteration
- partially observable
- multi agent
- decision theoretic
- supervised learning
- expected utility
- computational complexity
- control strategies
- average cost
- infinite horizon
- control policy
- real valued