On the Power and Limitations of Random Features for Understanding Neural Networks.
Gilad YehudaiOhad ShamirPublished in: CoRR (2019)
Keyphrases
- neural network
- feature extraction
- pattern recognition
- artificial neural networks
- feature vectors
- classification accuracy
- power consumption
- genetic algorithm
- feature space
- database
- benchmark datasets
- relative strengths and weaknesses
- special features
- salient features
- neural network model
- svm classifier
- false positives
- feature set
- co occurrence
- low level