Regularized nonsmooth Newton method for multi-class support vector machines

DOI10.1080/10556780600834745zbMath1188.68256OpenAlexW1995833382MaRDI QIDQ5758177

Ping Zhong, Masao Fukushima

Publication date: 3 September 2007

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

Full work available at URL: https://doi.org/10.1080/10556780600834745

zbMATH Keywords

regularization methodsmulti-class classificationD-gap functionbox constrained variational inequality problemnonsmooth Newton method

Mathematics Subject Classification ID

Computational learning theory (68Q32) Numerical mathematical programming methods (65K05) Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10)

Related Items (5)

AdaBoost Semiparametric Model Averaging Prediction for Multiple Categories ⋮ RT-GSOM: rough tolerance growing self-organizing map ⋮ An efficient augmented Lagrangian method for support vector machine ⋮ Spike-Timing Error Backpropagation in Theta Neuron Networks ⋮ A Newton-like method for solving a non-smooth elliptic equation

Uses Software

• Matlab
• UCI-ml
• RSVM

Cites Work

• Unnamed Item
• A note on least two norm solutions of monotone complementarity problems
• Equivalent differentiable optimization problems and descent methods for asymmetric variational inequality problems
• Solving box constrained variational inequalities by using the natural residual with D-gap function globalization
• Tikhonov regularization methods for variational inequality problems
• Unconstrained optimization reformulations of variational inequality problems
• 10.1162/15324430152733133
• 10.1162/15324430260185628
• Convergence Analysis of Some Algorithms for Solving Nonsmooth Equations
• A Nonmonotone Line Search Technique for Newton’s Method
• Finite-Dimensional Variational Inequalities and Complementarity Problems
• A new multi-class support vector algorithm

This page was built for publication: Regularized nonsmooth Newton method for multi-class support vector machines