EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Random vector functional link networks for function approximation on manifolds

Deanna Needell, Aaron A. Nelson, Rayan Saab, Palina Salanevich, Olov Schavemaker

Frontiers in Applied Mathematics and Statistics · 17.04.2024 · 10

DOI: 10.3389/fams.2024.1284706 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Applied Mathematics and Statistics, online erschienen am 17.04.2024. Verfasst von Deanna Needell und 4 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation, Simons Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The learning speed of feed-forward neural networks is notoriously slow and has presented a bottleneck in deep learning applications for several decades. For instance, gradient-based learning algorithms, which are used extensively to train neural networks, tend to work slowly when all of the network parameters must be iteratively tuned. To counter this, both researchers and practitioners have tried introducing randomness to reduce the learning requirement. Based on the original construction of Igelnik and Pao, single layer neural-networks with random input-to-hidden layer weights and biases have seen success in practice, but the necessary theoretical justification is lacking. In this study, we begin to fill this theoretical gap. We then extend this result to the non-asymptotic setting using a concentration inequality for Monte-Carlo integral approximations. We provide a (corrected) rigorous proof that the Igelnik and Pao construction is a universal approximator for continuous functions on compact domains, with approximation error squared decaying asymptotically like O (1/ n ) for the number n of network nodes. We then extend this result to the non-asymptotic setting, proving that one can achieve any desired approximation error with high probability provided n is sufficiently large. We further adapt this randomized neural network architecture to approximate functions on smooth, compact submanifolds of Euclidean space, providing theoretical guarantees in both the asymptotic and non-asymptotic forms. Finally, we illustrate our results on manifolds with numerical experiments.

Förderung

  • National Science Foundation
  • Simons Foundation

Zitieren

APA

Needell, D., Nelson, A. A., Saab, R., Salanevich, P., & Schavemaker, O. (2024). Random vector functional link networks for function approximation on manifolds. Frontiers in Applied Mathematics and Statistics, 10. https://doi.org/10.3389/fams.2024.1284706

Vancouver

Needell D, Nelson AA, Saab R, Salanevich P, Schavemaker O. Random vector functional link networks for function approximation on manifolds. Frontiers in Applied Mathematics and Statistics. 2024;10. doi:10.3389/fams.2024.1284706

BibTeX
@article{Needell2024,
  author = {Needell, Deanna and Nelson, Aaron A. and Saab, Rayan and Salanevich, Palina and Schavemaker, Olov},
  title = {Random vector functional link networks for function approximation on manifolds},
  journal = {Frontiers in Applied Mathematics and Statistics},
  year = {2024},
  volume = {10},
  doi = {10.3389/fams.2024.1284706}
}
Kostenlos starten →