Accelerating Online Reinforcement Learning with Offline Datasets.
Ashvin NairMurtaza DalalAbhishek GuptaSergey LevinePublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- real time
- online learning
- function approximation
- learning algorithm
- state space
- amazon mechanical turk
- temporal difference
- information systems
- training dataset
- policy search
- database
- dynamic programming
- benchmark datasets
- optimal policy
- learning problems
- machine learning
- optimal control
- model free
- data sets