Reward Informed Dreamer for Task Generalization in Reinforcement Learning.
Chengyang YingZhongkai HaoXinning ZhouHang SuSongming LiuJialian LiDong YanJun ZhuPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- state space
- function approximation
- reward function
- optimal policy
- machine learning
- learning algorithm
- dynamic programming
- partially observable environments
- model free
- markov decision processes
- multi agent
- supervised learning
- learning problems
- temporal difference
- partially observable
- eligibility traces
- learning environment
- decision trees
- average reward
- robotic control
- transfer learning
- learning process
- decision making
- learning agent
- data mining