Iteratively-Reweighted Least-Squares Fitting of Support Vector Machines: A Majorization-Minimization Algorithm Approach

Support vector machines (SVMs) are an important tool in modern data analysis. Traditionally, support vector machines have been fitted via quadratic programming, either using purpose-built or off-the-shelf algorithms. We present an alternative approach to SVM fitting via the majorization–minimization (MM) paradigm. Algorithms that are derived via MM algorithm constructions can be shown to monotonically decrease their objectives at each iteration, as well as be globally convergent to stationary points. We demonstrate the construction of iteratively-reweighted least-squares (IRLS) algorithms, via the MM paradigm, for SVM risk minimization problems involving the hinge, least-square, squared-hinge, and logistic losses, and 1-norm, 2-norm, and elastic net penalizations. Successful implementations of our algorithms are presented via some numerical examples. ∗HDN is at the Department of Mathematics and Statistics, La Trobe University, Bundoora Victoria, Australia 3086 (email: [email protected]). GJM is at the School of Mathematics and Statistics, University of Queensland, St. Lucia Queensland, Australia 4072. 1 ar X iv :1 70 5. 04 65 1v 1 [ st at .C O ] 1 2 M ay 2 01 7