EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Learning in Deep Radial Basis Function Networks

Fabian Wurzberger, Friedhelm Schwenker iD

Entropy · 26.04.2024 · 26 (5) · 368

DOI: 10.3390/e26050368 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 26.04.2024. Verfasst von Fabian Wurzberger und 1 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Neural Information Processing, Ulm University, James-Franck-Ring, 89081 Ulm, Germany. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Learning in neural networks with locally-tuned neuron models such as radial Basis Function (RBF) networks is often seen as instable, in particular when multi-layered architectures are used. Furthermore, universal approximation theorems for single-layered RBF networks are very well established; therefore, deeper architectures are theoretically not required. Consequently, RBFs are mostly used in a single-layered manner. However, deep neural networks have proven their effectiveness on many different tasks. In this paper, we show that deeper RBF architectures with multiple radial basis function layers can be designed together with efficient learning schemes. We introduce an initialization scheme for deep RBF networks based on k-means clustering and covariance estimation. We further show how to make use of convolutions to speed up the calculation of the Mahalanobis distance in a partially connected way, which is similar to the convolutional neural networks (CNNs). Finally, we evaluate our approach on image classification as well as speech emotion recognition tasks. Our results show that deep RBF networks perform very well, with comparable results to other deep neural network types, such as CNNs.

Schlagwörter

Stichwörter der Autor:innen

Mahalanobis distanceclassificationfunction approximationfunction interpolationpartially connected neural networksradial basis function networks

Autor:innen und Institutionen

  • Institute of Neural Information Processing, Ulm University, James-Franck-Ring, 89081 Ulm, Germany.

ORCID-Profile: Friedhelm Schwenker

Zitieren

APA

Wurzberger, F., & Schwenker, F. (2024). Learning in Deep Radial Basis Function Networks. Entropy, 26(5), 368. https://doi.org/10.3390/e26050368

Vancouver

Wurzberger F, Schwenker F. Learning in Deep Radial Basis Function Networks. Entropy. 2024;26(5):368. doi:10.3390/e26050368

BibTeX
@article{Wurzberger2024,
  author = {Wurzberger, Fabian and Schwenker, Friedhelm},
  title = {Learning in Deep Radial Basis Function Networks},
  journal = {Entropy},
  year = {2024},
  volume = {26},
  number = {5},
  pages = {368},
  doi = {10.3390/e26050368}
}
Kostenlos starten →