English

MKL-$L_{0/1}$-SVM

Machine Learning 2023-09-06 v3 Machine Learning

Abstract

This paper presents a Multiple Kernel Learning (abbreviated as MKL) framework for the Support Vector Machine (SVM) with the (0,1)(0, 1) loss function. Some KKT-like first-order optimality conditions are provided and then exploited to develop a fast ADMM algorithm to solve the nonsmooth nonconvex optimization problem. Numerical experiments on real data sets show that the performance of our MKL-L0/1L_{0/1}-SVM is comparable with the one of the leading approaches called SimpleMKL developed by Rakotomamonjy, Bach, Canu, and Grandvalet [Journal of Machine Learning Research, vol. 9, pp. 2491-2521, 2008].

Cite

@article{arxiv.2308.12016,
  title  = {MKL-$L_{0/1}$-SVM},
  author = {Bin Zhu and Yijie Shi},
  journal= {arXiv preprint arXiv:2308.12016},
  year   = {2023}
}

Comments

26 pages in the JMLR template, 3 figures, and 2 tables, submitted to the Journal of Machine Learning Research, with minor text overlap with arXiv: 2303.04445 (conference version). arXiv admin note: text overlap with arXiv:2303.04445

R2 v1 2026-06-28T12:02:20.365Z