Offline Reinforcement Learning with Soft Behavior Regularization.
Haoran XuXianyuan ZhanJianxiong LiHonglei YinPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- learning algorithm
- real time
- dynamic programming
- state space
- human behavior
- model free
- temporal difference
- reinforcement learning algorithms
- behavior analysis
- robotic control
- multi agent reinforcement learning
- temporal difference learning
- behavior patterns
- action selection
- regularization parameter
- optimal control
- image processing