In this paper we present ExSpliNet, an interpretable and expressive neural network model. The model combines ideas of Kolmogorov neural networks, ensembles of probabilistic trees, and multivariate B-spline representations. We give a probabilistic interpretation of the model and show its universal approximation properties. We also discuss how it can be efficiently encoded by exploiting B-spline properties. Finally, we test the effectiveness of the proposed model on synthetic approximation problems and classical machine learning benchmark datasets.

Fakhoury, D., Fakhoury, E., Speleers, H. (2022). ExSpliNet: an interpretable and expressive spline-based neural network. NEURAL NETWORKS, 152, 332-346 [10.1016/j.neunet.2022.04.029].

ExSpliNet: an interpretable and expressive spline-based neural network

Speleers H.
2022-08-01

Abstract

In this paper we present ExSpliNet, an interpretable and expressive neural network model. The model combines ideas of Kolmogorov neural networks, ensembles of probabilistic trees, and multivariate B-spline representations. We give a probabilistic interpretation of the model and show its universal approximation properties. We also discuss how it can be efficiently encoded by exploiting B-spline properties. Finally, we test the effectiveness of the proposed model on synthetic approximation problems and classical machine learning benchmark datasets.
ago-2022
Pubblicato
Rilevanza internazionale
Articolo
Esperti anonimi
Settore MAT/08 - ANALISI NUMERICA
English
Con Impact Factor ISI
Kolmogorov neural networks; Probabilistic trees; Tensor-product B-splines
Fakhoury, D., Fakhoury, E., Speleers, H. (2022). ExSpliNet: an interpretable and expressive spline-based neural network. NEURAL NETWORKS, 152, 332-346 [10.1016/j.neunet.2022.04.029].
Fakhoury, D; Fakhoury, E; Speleers, H
Articolo su rivista
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2108/302790
Citazioni
  • ???jsp.display-item.citation.pmc??? 0
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 2
social impact