Small Dataset, Big Gains: Enhancing Reinforcement Learning by Offline Pre-Training with Model Based Augmentation.
Girolamo MacalusoAlessandro SestiniAndrew D. BagdanovPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- model free
- training dataset
- test set
- benchmark datasets
- training examples
- training set
- function approximation
- optimal control
- small number
- state space
- learning algorithm
- database
- real world
- real time
- synthetic datasets
- training and test sets
- supervised learning
- machine learning
- data driven
- training process
- artificial neural networks
- multi agent
- policy search