Variational oracle guiding for reinforcement learning.
Dongqi HanTadashi KozunoXufang LuoZhao-Yun ChenKenji DoyaYuqing YangDongsheng LiPublished in: ICLR (2022)
Keyphrases
- reinforcement learning
- function approximation
- image segmentation
- state space
- temporal difference
- model free
- database
- markov decision processes
- robotic control
- oracle database
- machine learning
- reinforcement learning algorithms
- variational methods
- transition model
- methods in computer vision
- dynamic programming
- optical flow
- multi agent
- reinforcement learning methods
- optical flow computation
- framework for image segmentation
- action selection
- optimal control
- optimal policy
- scale space
- learning process
- databases