Incremental Reinforcement Learning via Performance Evaluation and Policy Perturbation.
Guizhou DengHuiqiao FuXinpeng WangCanghai LiuKaiqiang TangChunlin ChenPublished in: IEEE Trans. Artif. Intell. (2024)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- action selection
- partially observable environments
- actor critic
- control policy
- action space
- state and action spaces
- state space
- reinforcement learning problems
- markov decision processes
- function approximation
- partially observable markov decision processes
- policy iteration
- reward function
- control policies
- function approximators
- policy gradient
- partially observable
- model free
- dynamic programming
- policy evaluation
- markov decision problems
- state action
- control problems
- multi agent
- incremental learning
- infinite horizon
- reinforcement learning algorithms
- continuous state spaces
- decision problems
- rl algorithms
- temporal difference
- approximate dynamic programming
- learning algorithm
- average reward
- long run
- learning process
- machine learning
- policy making
- batch mode
- transfer learning
- agent learns
- robotic control
- approximate policy iteration
- partially observable domains