A novel action decision method of deep reinforcement learning based on a neural network and confidence bound.
Wenhao ZhangYaqing SongXiangpeng LiuQianqian ShangguanKang AnPublished in: Appl. Intell. (2023)
Keyphrases
- neural network
- high accuracy
- reinforcement learning
- significant improvement
- decision rules
- lower bound
- preprocessing
- computational cost
- upper bound
- edge detection
- detection method
- objective function
- computational complexity
- cost function
- decision making
- clustering method
- neural network is trained
- training algorithm
- confidence measure
- support vector machine svm
- support vector machine
- dynamic programming
- similarity measure
- machine learning