CLUE: Calibrated Latent Guidance for Offline Reinforcement Learning.
Jinxin LiuLipeng ZuLi HeDonglin WangPublished in: CoRL (2023)
Keyphrases
- reinforcement learning
- function approximation
- real time
- model free
- latent variables
- markov decision processes
- state space
- data sets
- optimal policy
- learning process
- reinforcement learning algorithms
- neural network
- learning algorithm
- multi agent
- robotic control
- dynamic programming
- learning problems
- transfer learning
- probabilistic model
- e learning
- video sequences
- multi view
- action selection
- supervised learning
- function approximators
- latent variable models
- autonomous learning
- transition model
- policy search