A P-ADMM for sparse quadratic kernel-free least squares semi-supervised support vector machine.
Yaru ZhanYanqin BaiWei ZhangShihui YingPublished in: Neurocomputing (2018)
Keyphrases
- least squares
- semi supervised
- support vector machine
- kernel function
- multiple kernel
- sparse linear
- sparse kernel
- kernel methods
- kernel machines
- pairwise
- support vector
- svm classifier
- semi supervised learning
- kernel matrix
- canonical correlation analysis
- low rank
- feature space
- kernel learning
- reduced set
- multi class
- additive models
- svm classification
- robust estimation
- unlabeled data
- multiple kernel learning
- feature vectors
- labeled data
- feature selection
- support vector machine svm
- supervised learning
- ls svm
- active learning
- high dimensional
- radial basis function
- machine learning
- training data
- total variation
- pairwise constraints
- reproducing kernel hilbert space
- kernel principal component analysis
- compressed sensing
- semi supervised clustering
- learning algorithm
- metric learning
- hyperplane
- convex optimization
- singular value decomposition
- objective function
- knn
- optical flow
- training set
- natural images
- histogram intersection kernel
- alternating direction method of multipliers
- image processing
- feature extraction
- regularized least squares
- decision boundary
- training samples
- sparse representation
- sparse coding
- support vectors