Q-value Regularized Transformer for Offline Reinforcement Learning.
Shengchao HuZiqing FanChaoqin HuangLi ShenYa ZhangYanfeng WangDacheng TaoPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- least squares
- fuzzy logic
- state space
- fault diagnosis
- machine learning
- learning algorithm
- reinforcement learning algorithms
- total least squares
- supervised learning
- regularized least squares
- real time
- regularization method
- markov decision processes
- power transformers
- robotic control
- incipient fault
- partially observable
- learning capabilities
- robot control
- markov decision process
- computational intelligence
- risk minimization
- autonomous learning
- evolutionary algorithm
- multi agent
- objective function
- real world