8:00 AM-8:45 AM
Room: Capitol North/Center South
Chair: Michael C. Ferris, University of Wisconsin, Madison
Optimization theory and algorithms have played a significant role in a great variety of fields. In this presentation, contributions of mathematical programming to the following problems of machine learning and data mining will be highlighted: linear and nonlinear separation of patterns, feature selection in discrimination problems, misclassification minimization, clustering problems, parsimonious solution of linear systems, support vector machines and massive data discrimination. These and other problems can be formulated and solved as linear or quadratic programs, or as concave function minimization on a polytope.
Olvi L. Mangasarian
Department of Computer Science
University of Wisconsin, Madison