Structured Variable Selection for Regularized Generalized Canonical Correlation Analysis, The Multiple Facets of Partial Least Squares and Related Methods - Archive ouverte HAL Access content directly
Book Sections Year : 2016

Structured Variable Selection for Regularized Generalized Canonical Correlation Analysis, The Multiple Facets of Partial Least Squares and Related Methods

(1) , (1) , (1) , (2) , (3) , (1) , (4)
1
2
3
4
Tommy Lofstedt
  • Function : Author
Fouad Hadj-Selem
  • Function : Author
  • PersonId : 950460
Vincent Guillemot
  • Function : Author
  • PersonId : 854230
Vincent Frouin
Arthur Tenenhaus

Abstract

Regularized Generalized Canonical Correlation Analysis (RGCCA) extends regularized canonical correlation analysis to more than two sets of variables. Sparse GCCA (SGCCA) was recently proposed to address the issue of variable selection. However, the variable selection scheme offered by SGCCA is limited to the covariance (τ = 1) link between blocks. In this paper we go beyond the covariance link by proposing an extension of SGCCA for the full RGCCA model (τ ∈ [0, 1]). In addition, we also propose an extension of SGCCA that exploits pre-given structural relationships between variables within blocks. Specifically, we propose an algorithm that allows structured and sparsity-inducing penalties to be included in the RGCCA optimization problem.
Not file

Dates and versions

hal-01396614 , version 1 (14-11-2016)

Identifiers

Cite

Tommy Lofstedt, Fouad Hadj-Selem, Vincent Guillemot, Cathy Philippe, Edouard Duchesnay, et al.. Structured Variable Selection for Regularized Generalized Canonical Correlation Analysis, The Multiple Facets of Partial Least Squares and Related Methods . Springer Proceedings in Mathematics & Statistics, pp.129-139, 2016, The Multiple Facets of Partial Least Squares and Related Methods, ⟨10.1007/978-3-319-40643-5_10⟩. ⟨hal-01396614⟩
274 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More