A finite newton method for classification

DOI10.1080/1055678021000028375zbMath1065.90078MaRDI QIDQ4405946

Olvi L. Mangasarian

Publication date: 2002

Published in: Optimization Methods and Software (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1080/1055678021000028375

zbMATH Keywords

Newton Method, Classification, Data Mining, Support Vector Machines

Mathematics Subject Classification ID

65K10: Numerical optimization and variational techniques

90C20: Quadratic programming

90C53: Methods of quasi-Newton type

Related Items

Training a Support Vector Machine in the Primal, Armijo Newton method for convex best interpolation, Augmented Lagrangian method for large-scale linear programming problems, A global piecewise smooth Newton method for fast large-scale model predictive control, A Newton method for linear programming, Finding the projection of a given point on the set of solutions of a linear programming problem, Regularity and well-posedness of a dual program for convex best \(C^{1}\)-spline interpolation, Unnamed Item, Binary separation and training support vector machines, Training robust support vector regression with smooth non-convex loss function, Exactness conditions for a convex differentiable exterior penalty for linear programming, Recursive Finite Newton Algorithm for Support Vector Regression in the Primal, Chunking for massive nonlinear kernel classification

Uses Software

• QPBOX
• SSVM

Cites Work

• Generalized Hessian matrix and second-order optimality conditions for problems with \(C^{1,1}\) data
• Bound constrained quadratic programming via piecewise quadratic functions
• Superlinearly convergent approximate Newton methods for LC\(^ 1\) optimization problems
• Minimization of \(SC^ 1\) functions and the Maratos effect
• A nonsmooth version of Newton's method
• Minimization of functions having Lipschitz continuous first partial derivatives
• A Newton Method for Convex Regression, Data Smoothing, and Quadratic Programming with Bounded Constraints
• On second-order sufficient optimality conditions for c ^1,1-optimization problems
• A Finite Continuation Algorithm for Bound Constrained Quadratic Programming
• A New Algorithm for Solving Strictly Convex Quadratic Programs
• Refinements of necessary optimality conditions in nondifferentiable programming II
• Multicategory proximal support vector machine classifiers
• SSVM: A smooth support vector machine for classification
• Unnamed Item
• Unnamed Item
• Unnamed Item