Stable deep reinforcement learning method by predicting uncertainty in rewards as a subtask.
Kanata SuzukiTetsuya OgataPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- high accuracy
- neural network
- computational complexity
- clustering method
- detection method
- similarity measure
- pairwise
- experimental evaluation
- planning problems
- data sets
- markov decision processes
- domain independent
- high precision
- segmentation method
- missing data
- edge detection
- state space
- dynamic programming
- cost function
- multiscale
- image segmentation