Domain adaptive state representation alignment for reinforcement learning.
Dongfen LiLichao MengJingjing LiKe LuYang YangPublished in: Inf. Sci. (2022)
Keyphrases
- reinforcement learning
- state space
- hidden state
- temporal abstractions
- domain specific
- image representation
- domain independent
- explicitly represent
- machine learning
- actor critic
- representation scheme
- cross domain
- state variables
- function approximation
- learning problems
- mobile robot
- dynamic programming
- learning process
- multi agent
- learning algorithm