-norm for high level classification features pruning in modern convolutional neural networks with local sensitivity analysis.
Ernest JeczmionekPiotr A. KowalskiPublished in: Int. J. Appl. Math. Comput. Sci. (2023)
Keyphrases
- sensitivity analysis
- high level
- low level
- feature vectors
- classification models
- feature set
- classification accuracy
- classification method
- feature extraction
- managerial insights
- classification process
- svm classifier
- feature space
- convolutional neural networks
- feature analysis
- support vector machine svm
- influence diagrams
- class labels
- benchmark datasets
- decision trees
- machine learning
- text classification
- variational inequalities
- extracted features
- training set
- feature selection
- classification algorithm
- image classification
- supervised learning
- support vector
- feature subset
- decision problems
- image sequences
- convolutional network