Offline Reinforcement Learning with Policy Guidance and Uncertainty Estimation.
Lan WuQuan LiuLihua ZhangZhigang HuangPublished in: ICASSP (2024)
Keyphrases
- reinforcement learning
- optimal policy
- sequential decision making
- policy search
- action selection
- control policy
- partially observable environments
- partial observability
- function approximation
- approximate dynamic programming
- markov decision process
- markov decision processes
- state and action spaces
- estimation algorithm
- decision problems
- partially observable
- dynamic programming
- policy evaluation
- markov decision problems
- agent receives
- neural network
- action space
- reinforcement learning algorithms
- reward function
- machine learning
- policy iteration
- rl algorithms
- control policies
- accurate estimation
- continuous state
- decision theory
- learning problems
- sufficient conditions
- reinforcement learning problems
- least squares
- state space