Deep Support Vector Classification and Regression

Abstract

Support Vector Machines, SVM, are one of the most popular machine learning models for supervised problems and have proved to achieve great performance in a wide broad of predicting tasks. However, they can suffer from scalability issues when working with large sample sizes, a common situation in the big data era. On the other hand, Deep Neural Networks (DNNs) can handle large datasets with greater ease and in this paper we propase Deep SVM models that combine the highly non-linear feature processing of DNNs with SVM loss functions. As we will show, these models can achieve performances similar to those of standard SVM while having a greater sample scalability.

Artificial intelligence within the interplay between natural and artificial computation: Advances in data science, trends and applications

“…Although some proposals attempt to overcome this obstacle, in many cases they still struggle with large datasets. The developments in [26] describe deep ANN models for both classification and regression which leverage the loss functions of support vector machines and support vector regression models, respectively. The results show that the performance is very similar to that of the traditional models, but the training time required to learn an ANN-based model is much shorter.…”

“…An example of research to complement certain models and trying to solve some of their problems can be found in [26]. While support vector machines (SVM) are the most successful models for supervised learning problems, it is true that they suffer from scalability problems with large amounts of data.…”