Back to Search
Start Over
A survey of numerical algorithms that can solve the Lasso problems.
- Source :
-
WIREs: Computational Statistics . Jul/Aug2023, Vol. 15 Issue 4, p1-15. 15p. - Publication Year :
- 2023
-
Abstract
- In statistics, the least absolute shrinkage and selection operator (Lasso) is a regression method that performs both variable selection and regularization. There is a lot of literature available, discussing the statistical properties of the regression coefficients estimated by the Lasso method. However, there lacks a comprehensive review discussing the algorithms to solve the optimization problem in Lasso. In this review, we summarize five representative algorithms to optimize the objective function in Lasso, including iterative shrinkage threshold algorithm (ISTA), fast iterative shrinkageâthresholding algorithms (FISTA), coordinate gradient descent algorithm (CGDA), smooth L1 algorithm (SLA), and path following algorithm (PFA). Additionally, we also compare their convergence rate, as well as their potential strengths and weakness. This article is categorized under:Statistical Models > Linear ModelsAlgorithms and Computational Methods > Numerical MethodsAlgorithms and Computational Methods > Computational Complexity [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 19395108
- Volume :
- 15
- Issue :
- 4
- Database :
- Academic Search Index
- Journal :
- WIREs: Computational Statistics
- Publication Type :
- Academic Journal
- Accession number :
- 165469663
- Full Text :
- https://doi.org/10.1002/wics.1602