Training very large scale nonlinear SVMs using Alternating Direction Method of Multipliers coupled with the Hierarchically Semi-Separable kernel approximations.
Stefano CipollaJacek GondzioPublished in: EURO J. Comput. Optim. (2022)
Keyphrases
- support vector
- linear svm
- kernel function
- kernel principal component analysis
- kernel methods
- training set
- kernel pca
- alternating direction method of multipliers
- kernel svms
- training examples
- feature selection
- high dimensional feature space
- hyperplane
- feature space
- multi class
- text categorization
- training process
- support vectors
- kernel matrix
- stochastic gradient descent
- input space
- multiple kernel learning
- training samples
- training data
- svm classifier
- training dataset
- support vector machine
- matrix completion
- learning algorithm