Kernel selection with spectral perturbation stability of kernel matrix.
Yong LiuShizhong LiaoPublished in: Sci. China Inf. Sci. (2014)
Keyphrases
- kernel matrix
- kernel methods
- kernel function
- low rank
- semidefinite programming
- kernel learning
- feature space
- positive definite
- kernel matrices
- kernel pca
- model selection
- positive semidefinite
- input space
- metric learning
- reproducing kernel hilbert space
- support vectors
- training samples
- kernel ridge regression
- semi definite programming
- ls svm
- linear combination
- support vector machine
- eigendecomposition
- support vector
- least squares
- semi supervised
- learning problems
- learning tasks
- dimensionality reduction
- knn
- training set
- training data