Speeding up reinforcement learning by combining attention and agency features.
Berkay DemirelMartí Sánchez-FiblaPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- feature set
- neural network
- feature vectors
- function approximation
- low level
- extracted features
- classification accuracy
- image features
- co occurrence
- benchmark datasets
- model free
- salient features
- structural information
- classification method
- spatial information
- dynamic programming
- computer vision