Efficient Training of One Class Classification-SVMs

TL;DR

This paper introduces an efficient training algorithm for one-class SVMs using the Augmented Lagrangian Fast Projected Gradient Method, enabling scalable and effective classification when only positive examples are available.

Contribution

It presents a novel, computationally inexpensive algorithm for training dual soft-margin one-class SVMs, suitable for large datasets and complementing existing solvers.

Findings

The proposed method achieves statistically significant results on real-world datasets.

It requires only first derivatives, reducing computational complexity.

The approach is scalable and effective for large-scale one-class classification.

Abstract

This study examines the use of a highly effective training method to conduct one-class classification. The existence of both positive and negative examples in the training data is necessary to develop an effective classifier in common binary classification scenarios. Unfortunately, this criteria is not met in many domains. Here, there is just one class of examples. Classification algorithms that learn from solely positive input have been created to deal with this setting. In this paper, an effective algorithm for dual soft-margin one-class SVM training is presented. Our approach makes use of the Augmented Lagrangian (AL-FPGM), a variant of the Fast Projected Gradient Method. The FPGM requires only first derivatives, which for the dual soft margin OCC-SVM means computing mainly a matrix-vector product. Therefore, AL-FPGM, being computationally inexpensive, may complement existing quadratic programming solvers for training large SVMs. We extensively validate our approach over real-world datasets and demonstrate that our strategy obtains statistically significant results.