Regret Bounds for Risk-Sensitive Reinforcement Learning.
Osbert BastaniYecheng Jason MaEstelle ShenWanqiao XuPublished in: NeurIPS (2022)
Keyphrases
- risk sensitive
- reinforcement learning
- model free
- optimal control
- markov decision processes
- regret bounds
- control policies
- markov decision problems
- reinforcement learning algorithms
- function approximation
- utility function
- optimal policy
- temporal difference
- state space
- average cost
- policy iteration
- linear regression
- online learning
- machine learning
- dynamic programming
- markov decision process
- finite state
- learning algorithm
- infinite horizon
- transfer learning
- action space
- learning process
- partially observable
- reward function
- average reward
- lower bound
- computational complexity