Using the Ellipsoid Method to Find Parameters of Lasso and Ridge Regressions [Articol]
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
We consider the optimization problem for finding the parameters of a linear regression according to the criterion of the least moduli powered to p (1 ≤ p ≤ 2) with the regularization of parameters according to the criterion of the least moduli powered to q (1 ≤ q ≤ 2). Its partial cases are lasso regression and ridge regression, as well as least squares method and the least moduli method. An algorithm for solving the problem is developed based on the well-known ellipsoid method.
Description
Keywords
Lasso regression, ridge regression, linear regres- sion, least moduli criterion, convex function, ellipsoid method
Citation
STETSYUK, Petro and Olha KHOMIAK. Using the Ellipsoid Method to Find Parameters of Lasso and Ridge Regressions. In: International Conference dedicated to the 60th anniversary of the foundation of Vladimir Andrunachievici Institute of Mathematics and Computer Science, MSU, October 10-13 2024. Chisinau: [S. n.], 2024, pp. 472-475. ISBN 978-9975-68-515-3.