EUVIMEDEUVIMED entdecken
Crossref

Interpolation, Approximation, and Controllability of Deep Neural Networks

Jingpu Cheng, Qianxiao Li, Ting Lin, Zuowei Shen

SIAM Journal on Control and Optimization · 28.02.2025 · 63 (1) · 625-649

DOI: 10.1137/23m1599744 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SIAM Journal on Control and Optimization, online erschienen am 18.02.2025. Verfasst von Jingpu Cheng und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: National Research Foundation, National University of Singapore.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We investigate the expressive power of deep residual neural networks idealized as continuous dynamical systems through control theory. Specifically, we consider two properties that arise from supervised learning, namely universal interpolation—the ability to match arbitrary input and target training samples—and the closely related notion of universal approximation—the ability to approximate input-target functional relationships via flow maps. Under the assumption of affine invariance of the control family, we give a characterization of universal interpolation, showing that it holds for essentially any architecture with nonlinearity. Furthermore, we elucidate the relationship between universal interpolation and universal approximation in the context of general control systems, showing that the two properties cannot be deduced from each other. At the same time, we identify conditions on the control family and the target function that ensure the equivalence of the two notions.

Förderung

  • National Research Foundation
  • National University of Singapore

Zitieren

APA

Cheng, J., Li, Q., Lin, T., & Shen, Z. (2025). Interpolation, Approximation, and Controllability of Deep Neural Networks. SIAM Journal on Control and Optimization, 63(1), 625-649. https://doi.org/10.1137/23m1599744

Vancouver

Cheng J, Li Q, Lin T, Shen Z. Interpolation, Approximation, and Controllability of Deep Neural Networks. SIAM Journal on Control and Optimization. 2025;63(1):625-649. doi:10.1137/23m1599744

BibTeX
@article{Cheng2025,
  author = {Cheng, Jingpu and Li, Qianxiao and Lin, Ting and Shen, Zuowei},
  title = {Interpolation, Approximation, and Controllability of Deep Neural Networks},
  journal = {SIAM Journal on Control and Optimization},
  year = {2025},
  volume = {63},
  number = {1},
  pages = {625--649},
  doi = {10.1137/23m1599744}
}
Kostenlos starten →