Offline Reinforcement Learning with Reverse Model-based Imagination.
Jianhao WangWenzhe LiHaozhe JiangGuangxiang ZhuSiyuan LiChongjie ZhangPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- model free
- state space
- real time
- function approximation
- reinforcement learning algorithms
- learning capabilities
- machine learning
- partially observable
- optimal policy
- markov decision processes
- direct policy search
- robotic control
- policy search
- temporal difference
- expert systems
- case study
- knowledge base
- real world
- neural network
- data sets