Efficient methods for robust classification under uncertainty in kernel matrices

A. Ben-Tal, S. Bhadra, C. Bhattacharyya, A. Nemirovski

    Research output: Contribution to journalArticleScientificpeer-review

    13 Citations (Scopus)


    In this paper we study the problem of designing SVM classifiers when the kernel matrix, K, is affected by uncertainty. Specifically K is modeled as a positive affine combination of given positive semi definite kernels, with the coefficients ranging in a norm-bounded uncertainty set. We treat the problem using the Robust Optimization methodology. This reduces the uncertain SVM problem into a deterministic conic quadratic problem which can be solved in principle by a polynomial time Interior Point (IP) algorithm. However, for large-scale classification problems, IP methods become intractable and one has to resort to first-order gradient type methods. The strategy we use here is to reformulate the robust counterpart of the uncertain SVM problem as a saddle point problem and employ a special gradient scheme which works directly on the convex-concave saddle function. The algorithm is a simplified version of a general scheme due to Juditski and Nemirovski (2011). It achieves an O(1/T2) reduction of the initial error after T iterations. A comprehensive empirical study on both synthetic data and real-world protein structure data sets show that the proposed formulations achieve the desired robustness, and the saddle point based algorithm outperforms the IP method significantly.
    Original languageEnglish
    Pages (from-to)2923-2954
    JournalJournal of Machine Learning Research
    Publication statusPublished - 2012


    Dive into the research topics of 'Efficient methods for robust classification under uncertainty in kernel matrices'. Together they form a unique fingerprint.

    Cite this