A Simple Method for Solving the SVM Regularization Path for Semidefinite Kernels.
Christopher SentelleGeorgios C. AnagnostopoulosMichael GeorgiopoulosPublished in: IEEE Trans. Neural Networks Learn. Syst. (2016)
Keyphrases
- semidefinite
- positive semidefinite
- polynomial kernels
- support vector
- kernel function
- semidefinite programming
- multiple kernel learning
- kernel methods
- convex relaxation
- support vector machine svm
- interior point methods
- quadratic programming
- reproducing kernel hilbert space
- gaussian kernels
- support vector machine
- feature space
- svm classifier
- sufficient conditions
- knn
- higher dimensional
- maximum margin
- training data
- string kernels
- convex sets
- kernel matrix
- cross validation
- linear combination
- feature selection
- generalization ability
- finite dimensional
- loss function
- dynamic programming
- np hard
- feature vectors