Value-aware Importance Weighting for Off-policy Reinforcement Learning.
Kristopher De AsisEric GravesRichard S. SuttonPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- optimal control
- feature weighting
- databases
- learning algorithm
- state space
- optimal policy
- model free
- learning process
- relative importance
- transfer learning
- function approximation
- reinforcement learning algorithms
- robotic control
- learning capabilities
- weighting schemes
- database
- weighting scheme
- probabilistic model
- dynamic programming
- machine learning
- data sets