Distance Metric Learning for Kernel Machines
This addresses the problem of enhancing kernel machine performance for researchers and practitioners in machine learning, though it is incremental as it builds on existing metric learning and SVM methods.
The paper tackled the problem of improving SVM-RBF classification by evaluating existing Mahalanobis metric learning algorithms and found they did not yield significant improvements. It introduced SVML, a novel algorithm that combines metric learning with SVM training, which outperformed state-of-the-art alternatives on nine benchmark datasets in terms of accuracy.
Recent work in metric learning has significantly improved the state-of-the-art in k-nearest neighbor classification. Support vector machines (SVM), particularly with RBF kernels, are amongst the most popular classification algorithms that uses distance metrics to compare examples. This paper provides an empirical analysis of the efficacy of three of the most popular Mahalanobis metric learning algorithms as pre-processing for SVM training. We show that none of these algorithms generate metrics that lead to particularly satisfying improvements for SVM-RBF classification. As a remedy we introduce support vector metric learning (SVML), a novel algorithm that seamlessly combines the learning of a Mahalanobis metric with the training of the RBF-SVM parameters. We demonstrate the capabilities of SVML on nine benchmark data sets of varying sizes and difficulties. In our study, SVML outperforms all alternative state-of-the-art metric learning algorithms in terms of accuracy and establishes itself as a serious alternative to the standard Euclidean metric with model selection by cross validation.