A simple yet efficient algorithm for multiple kernel learning under elastic-net constraints

TL;DR

This paper presents a straightforward and efficient algorithm for multiple kernel learning with elastic-net constraints, offering significant improvements in computational efficiency and simplicity over existing methods.

Contribution

The paper introduces a novel, simple algorithm for MKL with elastic-net constraints that outperforms existing approaches in speed and memory usage.

Findings

Favorable comparison in time and space complexity

Easy to implement without external libraries

Compatible with standard SVM solvers

Abstract

This papers introduces an algorithm for the solution of multiple kernel learning (MKL) problems with elastic-net constraints on the kernel weights. The algorithm compares very favourably in terms of time and space complexity to existing approaches and can be implemented with simple code that does not rely on external libraries (except a conventional SVM solver).