Provable Reward-Agnostic Preference-Based Reinforcement Learning.
Wenhao ZhanMasatoshi UeharaWen SunJason D. LeePublished in: ICLR (2024)
Keyphrases
- reinforcement learning
- state space
- function approximation
- temporal difference learning
- learning algorithm
- markov decision processes
- reinforcement learning methods
- reinforcement learning algorithms
- eligibility traces
- partially observable environments
- optimal policy
- multi agent
- user preferences
- policy search
- temporal difference
- robotic control
- reward function
- least squares
- control policy
- model free
- average reward
- case based problem solving
- optimal control
- evaluation function
- learning problems
- transfer learning
- supervised learning