ConserWeightive Behavioral Cloning for Reliable Offline Reinforcement Learning.
Tung NguyenQinqing ZhengAditya GroverPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- machine learning
- optimal policy
- selective perception
- reinforcement learning algorithms
- model free
- highly accurate
- function approximation
- learning capabilities
- real time
- state space
- dynamic programming
- data sets
- least squares
- monte carlo
- artificial neural networks
- optimal control
- multi agent
- temporal difference
- image sequences
- databases
- control problems
- markov decision process
- multi agent reinforcement learning
- behavioral data
- robotic control