Interpretable Neural Networks with BP-SOM

Weijters, Ton; van den Bosch, Antal; van den Herik, H. Jaap

doi:10.1007/BFB0026711

Interpretable Neural Networks with BP-SOM

Ton Weijters, Antal van den Bosch, H. Jaap van den Herik

ECML-PKDD 1998 pp. 406-411

doi:10.1007/BFB0026711 /ecmlpkdd/1998/weijters1998ecml-interpretable/

Abstract

Interpretation of models induced by artificial neural networks is often a difficult task. In this paper we focus on a relatively novel neural network architecture and learning algorithm, bp-som that offers possibilities to overcome this difficulty. It is shown that networks trained with BP-SOM show interesting regularities, in that hidden-unit activations become restricted to discrete values, and that the som part can be exploited for automatic rule extraction.

PDF ECML-PKDD Semantic Scholar

Cite

Text

Weijters et al. "Interpretable Neural Networks with BP-SOM." European Conference on Machine Learning, 1998. doi:10.1007/BFB0026711

Markdown

[Weijters et al. "Interpretable Neural Networks with BP-SOM." European Conference on Machine Learning, 1998.](https://mlanthology.org/ecmlpkdd/1998/weijters1998ecml-interpretable/) doi:10.1007/BFB0026711

BibTeX

@inproceedings{weijters1998ecml-interpretable,
  title     = {{Interpretable Neural Networks with BP-SOM}},
  author    = {Weijters, Ton and van den Bosch, Antal and van den Herik, H. Jaap},
  booktitle = {European Conference on Machine Learning},
  year      = {1998},
  pages     = {406-411},
  doi       = {10.1007/BFB0026711},
  url       = {https://mlanthology.org/ecmlpkdd/1998/weijters1998ecml-interpretable/}
}