The Gradient’s Limit of a Definable Family of Functions Admits a Variational Stratification
SIAM Journal on Optimization · 30.06.2026 · 36 (2) · 1075-1099
DOI: 10.1137/24m1638306 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Optimization, online erschienen am 22.06.2026. Verfasst von Sholom Schechtman. Die Arbeit nennt 51 Literaturangaben.
Abstract (Originalsprache)
It is well known that the convergence of a family of smooth functions does not imply the convergence of its gradients. In this work, we show that if the family is definable in an o-minimal structure (for instance semialgebraic, subanalytic, or any composition of the previous with exp, log), then the gradient’s limit admits a variational stratification and, under a uniform Lipschitz continuity assumption, is a conservative set-valued field in the sense introduced by Bolte and Pauwels. Immediate implications of this result on convergence guarantees of smoothing methods are discussed. The result is established in a general form, where the functions in the original family might be non–Lipschitz continuous, be vector-valued with the gradients replaced by their Clarke Jacobians, or be an arbitrary mapping satisfying a definable variational stratification. In passing, we investigate stability properties of definable variational stratifications and smoothing methods that produce definable functions, which might be of independent interest.
Zitieren
APA
Schechtman, S. (2026). The Gradient’s Limit of a Definable Family of Functions Admits a Variational Stratification. SIAM Journal on Optimization, 36(2), 1075-1099. https://doi.org/10.1137/24m1638306
Vancouver
Schechtman S. The Gradient’s Limit of a Definable Family of Functions Admits a Variational Stratification. SIAM Journal on Optimization. 2026;36(2):1075-1099. doi:10.1137/24m1638306
BibTeX
@article{Schechtman2026,
author = {Schechtman, Sholom},
title = {The Gradient’s Limit of a Definable Family of Functions Admits a Variational Stratification},
journal = {SIAM Journal on Optimization},
year = {2026},
volume = {36},
number = {2},
pages = {1075--1099},
doi = {10.1137/24m1638306}
}
EUVIMED