Learning the Subspace of Variation for Global Optimization of Functions with Low Effective Dimension
SIAM Journal on Optimization · 31.12.2026 · 36 (4) · 2500-2529
DOI: 10.1137/24m1636642 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Optimization, online erschienen am 10.11.2026. Verfasst von Coralia Cartis und 3 weiteren Autor:innen. Die Arbeit nennt 55 Literaturangaben. Gefördert von: Engineering and Physical Sciences Research Council, Alan Turing Institute, National Physical Laboratory.
Abstract (Originalsprache)
We propose an algorithmic framework that employs active subspace techniques for scalable global optimization of functions with low effective dimension (also referred to as low-rank functions). This proposal replaces the original high-dimensional problem by one or several lower-dimensional reduced subproblem(s), capturing the main directions of variation of the objective, which are estimated here as the principal components of a collection of sampled gradients. We quantify the sampling complexity of estimating the subspace of variation of the objective in terms of its effective dimension and hence bound the probability that the reduced problem will provide a solution to the original problem. To account for the practical case when the effective dimension is not known a priori, our framework adaptively solves a sequence of reduced problems, increasing the number of sampled gradients until the estimated subspace of variation remains unchanged. We prove global convergence under mild assumptions on the objective, the sampling distribution, and the subproblem solver and illustrate numerically the benefits of our proposed algorithms over those using random embeddings.
Förderung
- Engineering and Physical Sciences Research Council
- Alan Turing Institute
- National Physical Laboratory
- Fonds De La Recherche Scientifique - FNRS
- International Business Machines Corporation
- Nazarbayev University
- Hong Kong Innovation and Technology Commission
- Faculty Development Competitive Research Grant Program
Zitieren
APA
Cartis, C., Liang, X., Massart, E., & Otemissov, A. (2026). Learning the Subspace of Variation for Global Optimization of Functions with Low Effective Dimension. SIAM Journal on Optimization, 36(4), 2500-2529. https://doi.org/10.1137/24m1636642
Vancouver
Cartis C, Liang X, Massart E, Otemissov A. Learning the Subspace of Variation for Global Optimization of Functions with Low Effective Dimension. SIAM Journal on Optimization. 2026;36(4):2500-2529. doi:10.1137/24m1636642
BibTeX
@article{Cartis2026,
author = {Cartis, Coralia and Liang, Xinzhu and Massart, Estelle and Otemissov, Adilet},
title = {Learning the Subspace of Variation for Global Optimization of Functions with Low Effective Dimension},
journal = {SIAM Journal on Optimization},
year = {2026},
volume = {36},
number = {4},
pages = {2500--2529},
doi = {10.1137/24m1636642}
}
EUVIMED