Harnessing Mixed Offline Reinforcement Learning Datasets via Trajectory Weighting.
Zhang-Wei HongPulkit AgrawalRémi Tachet des CombesRomain LarochePublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- function approximation
- learning algorithm
- data sets
- uci machine learning repository
- collective intelligence
- benchmark datasets
- state space
- multi agent systems
- transfer learning
- markov decision processes
- raw data
- model free
- weighting scheme
- real time
- learning process
- machine learning
- real world
- neural network
- action selection
- reinforcement learning algorithms