A full-Newton step infeasible interior-point method based on a trigonometric kernel function without centering steps.
Behrouz KheirfamMasoumeh HaghighiPublished in: Numer. Algorithms (2020)
Keyphrases
- interior point methods
- kernel function
- convex optimization
- semidefinite programming
- kernel matrix
- support vector
- kernel methods
- linear program
- linear programming
- primal dual
- feature space
- quadratic programming
- support vector machine
- input space
- computationally intensive
- kernel learning
- high dimensional
- feature set
- solving problems
- neural network
- semidefinite
- linear systems
- support vectors
- hyperplane
- reproducing kernel hilbert space
- semi supervised
- decision trees
- machine learning