Multicategory ψ-Learning and Support Vector Machine: Computational Tools
- 1 March 2005
- journal article
- Published by Taylor & Francis Ltd in Journal of Computational and Graphical Statistics
- Vol. 14 (1), 219-236
- https://doi.org/10.1198/106186005x37238
Abstract
Many margin-based binary classification techniques such as support vector machine (SVM) and ψ-learning deliver high performance. An earlier article proposed a new multicategory ψ-learning methodology that shows great promise in generalization ability. However,ψ-learning is computationally difficult because it requires handling a nonconvex minimization problem. In this article, we propose two computational tools for multicategory ψ-learning. The first one is based on d.c. algorithms and solved by sequential quadratic programming, while the second one uses the outer approximation method, which yields the global minimizer via sequential concave minimization. Numerical examples show the proposed algorithms perform well.Keywords
This publication has 7 references indexed in Scilit:
- On ψ-LearningJournal of the American Statistical Association, 2003
- Comparison of Discrimination Methods for the Classification of Tumors Using Gene Expression DataJournal of the American Statistical Association, 2002
- On Covering Methods for D.C. OptimizationJournal of Global Optimization, 2000
- Improved Generalization Through Explicit Optimization of MarginsMachine Learning, 2000
- Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression MonitoringScience, 1999
- Solving a Class of Linearly Constrained Indefinite Quadratic Problems by D.C. AlgorithmsJournal of Global Optimization, 1997
- Some results on Tchebycheffian spline functionsJournal of Mathematical Analysis and Applications, 1971