Iterative regularization in classification via hinge loss diagonal descent.
Vassilis ApidopoulosTomaso A. PoggioLorenzo RosascoSilvia VillaPublished in: CoRR (2022)
Keyphrases
- hinge loss
- loss function
- support vector
- support vector machine
- decision trees
- feature space
- convex optimization
- decision boundary
- potential functions
- multi class
- feature extraction
- supervised learning
- reproducing kernel hilbert space
- risk minimization
- machine learning
- training set
- feature vectors
- training samples
- binary classification
- line search