Higher-Order Reduced Rank Regression
SIAM Journal on Matrix Analysis and Applications · 30.06.2026 · 47 (2) · 724-757
DOI: 10.1137/25m1739637 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Matrix Analysis and Applications, online erschienen am 15.06.2026. Verfasst von Leia Greenberg und 1 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Gefördert von: Israel Science Foundation.
Abstract (Originalsprache)
Reduced Rank Regression (RRR) is a widely used method for multiresponse regression. However, RRR assumes a linear relationship between features and responses. While linear models are useful and often provide a good approximation, many real-world problems involve more complex relationships that cannot be adequately captured by simple linear interactions. One way to model such relationships is via multilinear transformations. This paper introduces Higher-Order Reduced Rank Regression (HORRR), an extension of RRR that leverages multilinear transformations and as such is capable of capturing nonlinear interactions in multiresponse regression. HORRR employs tensor representations for the coefficients and a Tucker decomposition to impose multilinear rank constraints as regularization akin to the rank constraints in RRR. Encoding these constraints as a manifold allows us to use Riemannian optimization to solve this HORRR problems. We theoretically and empirically analyze the use of Riemannian optimization for solving HORRR problems.
Förderung
- Israel Science Foundation
Zitieren
APA
Greenberg, L., & Avron, H. (2026). Higher-Order Reduced Rank Regression. SIAM Journal on Matrix Analysis and Applications, 47(2), 724-757. https://doi.org/10.1137/25m1739637
Vancouver
Greenberg L, Avron H. Higher-Order Reduced Rank Regression. SIAM Journal on Matrix Analysis and Applications. 2026;47(2):724-757. doi:10.1137/25m1739637
BibTeX
@article{Greenberg2026,
author = {Greenberg, Leia and Avron, Haim},
title = {Higher-Order Reduced Rank Regression},
journal = {SIAM Journal on Matrix Analysis and Applications},
year = {2026},
volume = {47},
number = {2},
pages = {724--757},
doi = {10.1137/25m1739637}
}
EUVIMED