On the Effect of Pre-training for Transformer in Different Modality on Offline Reinforcement Learning.
Shiro TakagiPublished in: NeurIPS (2022)
Keyphrases
- reinforcement learning
- multi modal
- state space
- fuzzy logic
- function approximation
- supervised learning
- training process
- training set
- fault diagnosis
- test set
- medical images
- markov decision processes
- training phase
- training samples
- learning algorithm
- model free
- training examples
- temporal difference
- independent variables
- serious games
- learning problems
- power system
- machine learning
- hidden markov models
- genetic algorithm