Continuous Doubly Constrained Batch Reinforcement Learning.
Rasool FakoorJonas MuellerKavosh AsadiPratik ChaudhariAlexander J. SmolaPublished in: NeurIPS (2021)
Keyphrases
- reinforcement learning
- action space
- machine learning
- markov decision processes
- learning algorithm
- continuous state and action spaces
- reinforcement learning algorithms
- model free
- function approximation
- multi agent
- optimal policy
- continuous domains
- batch mode
- supervised learning
- state space
- optimal control
- active learning
- temporal difference
- evolutionary algorithm
- robot control
- lower bound
- batch processing
- continuous state spaces
- batch size
- robotic control