Equivariant Offline Reinforcement Learning.
Arsh TangriOndrej BizaDian WangDavid KleeOwen HowellRobert PlattPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- real time
- markov decision processes
- reinforcement learning algorithms
- model free
- learning process
- control problems
- learning algorithm
- temporal difference
- dynamic programming
- action selection
- optimal control
- direct policy search
- optimal policy
- multi agent
- multiscale
- information retrieval
- data sets
- least squares
- real robot
- policy search
- robotic control