A Large Dimensional Analysis of Least Squares Support Vector Machines - Archive ouverte HAL Access content directly
Journal Articles IEEE Transactions on Signal Processing Year : 2019

A Large Dimensional Analysis of Least Squares Support Vector Machines

(1) , (1)
1
Zhenyu Liao
Romain Couillet

Abstract

In this article, a large dimensional performance analysis of kernel least squares support vector machines (LS-SVMs) is provided under the assumption of a two-class Gaussian mixture model for the input data. Building upon recent advances in random matrix theory, we show, when the dimension of data $p$ and their number $n$ are both large, that the LS-SVM decision function can be well approximated by a normally distributed random variable, the mean and variance of which depend explicitly on a local behavior of the kernel function. This theoretical result is then applied to the MNIST and Fashion-MNIST datasets which, despite their non-Gaussianity, exhibit a convincingly close behavior. Most importantly, our analysis provides a deeper understanding of the mechanism into play in SVM-type methods and in particular of the impact on the choice of the kernel function as well as some of their theoretical limits in separating high dimensional Gaussian vectors.
Fichier principal
Vignette du fichier
couillet_squares.pdf (375.19 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02048984 , version 1 (19-05-2020)

Identifiers

Cite

Zhenyu Liao, Romain Couillet. A Large Dimensional Analysis of Least Squares Support Vector Machines. IEEE Transactions on Signal Processing, 2019, 67 (4), pp.1065-1074. ⟨10.1109/TSP.2018.2889954⟩. ⟨hal-02048984⟩
77 View
184 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More