Frank-Wolfe Algorithm for Learning SVM-Type Multi-Category Classifiers

The multi-category support vector machine (MC-SVM) is one of the most popular learning algorithms. There are numerous MC-SVM variants, although different optimization algorithms were developed for diverse machines. In this study, we a new algorithm that can be applied to several variants. based on Frank-Wolfe framework requires two subproblems, direction-finding and line search, in each iteration. contribution study discovery both subproblems have closed form solution if dual problem. Additionally, solutions search exist even Moreau envelopes loss functions. We used large datasets demonstrate proposed rapidly converges thereby improves pattern recognition performance.