Very Fast Kernel SVM under Budget Constraints

TL;DR

This paper introduces a fast online Kernel SVM method that uses clustering and support vector set management to achieve high accuracy and processing speed under strict budget constraints.

Contribution

It presents a novel online Kernel SVM approach that employs LVQ-based clustering and support vector set size control for efficiency.

Findings

High accuracy achieved in experiments

Very high samples processed per second

Effective support vector set management

Abstract

In this paper we propose a fast online Kernel SVM algorithm under tight budget constraints. We propose to split the input space using LVQ and train a Kernel SVM in each cluster. To allow for online training, we propose to limit the size of the support vector set of each cluster using different strategies. We show in the experiment that our algorithm is able to achieve high accuracy while having a very high number of samples processed per second both in training and in the evaluation.