Gradient Flows for L2 Support Vector Machine Training

TL;DR

This paper investigates training support vector machines for binary classification using a continuous time approach via systems of ordinary differential equations, potentially benefiting hardware implementations like analog or quantum computers.

Contribution

It introduces a novel continuous time framework for SVM training, bridging machine learning and differential equations, suitable for emerging hardware platforms.

Findings

Proposes a continuous time perspective on SVM training.

Highlights potential for implementation on analog and quantum hardware.

Lays groundwork for future hardware-efficient SVM algorithms.

Abstract

We explore the merits of training of support vector machines for binary classification by means of solving systems of ordinary differential equations. We thus assume a continuous time perspective on a machine learning problem which may be of interest for implementations on (re)emerging hardware platforms such as analog- or quantum computers.