Optimized cutting plane algorithm for support vector machines
Vojtech Franc and Sören Sonnenburg
Fraunhofer Institute, Berlin, Germany.
We have developed a new Linear Support Vector Machine (SVM) training algorithm called OCAS, which itself significantly outperforms current state of the art SVM solvers, like svmlight, svmperf and BMRM, achieving speedups of over 1,200 on some datasets over svmlight and 29 over svmperf, while obtaining the same precise Support Vector solution. Using OCAS we were able to train on a dataset of size 15 million examples (itself about 32GB in size) in just 671 seconds - a competing string kernel SVM required 97,484 seconds to train on 10 million examples sub-sampled from this dataset. This was possible due to effective parallelizing of the core parts of OCAS which lead to additional speedups of factors up to 4.6 on a multi-core multiprocessor machine.