About the non-convex optimization problem induced by non-positive semidefinite kernel learning.
Ingo MierswaKatharina MorikPublished in: Adv. Data Anal. Classif. (2008)
Keyphrases
- convex optimization
- kernel matrix
- kernel learning
- positive semidefinite
- low rank
- semidefinite programming
- semidefinite program
- kernel function
- kernel matrices
- kernel methods
- interior point methods
- primal dual
- feature space
- convex relaxation
- convex sets
- support vectors
- total variation
- metric learning
- multiple kernel learning
- feature selection
- input space
- training samples
- support vector