Stable Deep Reinforcement Learning Method by Predicting Uncertainty in Rewards as a Subtask.
Kanata SuzukiTetsuya OgataPublished in: ICONIP (2) (2020)
Keyphrases
- reinforcement learning
- detection method
- clustering method
- computational cost
- model free
- similarity measure
- cost function
- classification accuracy
- high accuracy
- pairwise
- significant improvement
- experimental evaluation
- neural network
- mobile robot
- high precision
- segmentation method
- model selection
- support vector machine
- preprocessing
- image sequences
- machine learning