BRAC+: Improved Behavior Regularized Actor Critic for Offline Reinforcement Learning.
Chi ZhangSanmukh R. KuppannagariViktor K. PrasannaPublished in: ACML (2021)
Keyphrases
- actor critic
- reinforcement learning
- policy gradient
- temporal difference
- reinforcement learning algorithms
- optimal control
- function approximation
- approximate dynamic programming
- gradient method
- neuro fuzzy
- policy iteration
- state space
- policy gradient methods
- model free
- dynamic programming
- markov decision processes
- least squares
- rl algorithms
- step size
- optimal policy
- multi agent systems
- natural actor critic
- learning problems
- transfer learning
- supervised learning
- average reward
- temporal difference learning
- objective function
- machine learning