Fast model selection by limiting SVM training times

TL;DR

This paper proposes limiting training time during SVM parameter tuning to significantly reduce model selection duration without sacrificing performance.

Contribution

It introduces a novel stopping criterion for SVM training that accelerates model selection by an order of magnitude.

Findings

Training time reduced by an order of magnitude.

Model performance remains comparable.

Efficient parameter tuning process achieved.

Abstract

Kernelized Support Vector Machines (SVMs) are among the best performing supervised learning methods. But for optimal predictive performance, time-consuming parameter tuning is crucial, which impedes application. To tackle this problem, the classic model selection procedure based on grid-search and cross-validation was refined, e.g. by data subsampling and direct search heuristics. Here we focus on a different aspect, the stopping criterion for SVM training. We show that by limiting the training time given to the SVM solver during parameter tuning we can reduce model selection times by an order of magnitude.