Inelligible neural networks with BP-SOM

A.J.M.M. Weijters, A.P.J. Bosch, van den, H.J. Herik, van den

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

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.
Original languageEnglish
Title of host publicationMachine learning : ECML-98 : 10th European conference on machine learning, Chemnitz, Germany, April 21-23, 1998 : proceedings
EditorsC. Nédellec, C. Rouveirol
Place of PublicationBerlin
PublisherSpringer
Pages406-411
ISBN (Print)978-3-540-64417-0
DOIs
Publication statusPublished - 1998

Publication series

NameLecture notes in artificial intelligence
Volume1398

Fingerprint Dive into the research topics of 'Inelligible neural networks with BP-SOM'. Together they form a unique fingerprint.

  • Cite this

    Weijters, A. J. M. M., Bosch, van den, A. P. J., & Herik, van den, H. J. (1998). Inelligible neural networks with BP-SOM. In C. Nédellec, & C. Rouveirol (Eds.), Machine learning : ECML-98 : 10th European conference on machine learning, Chemnitz, Germany, April 21-23, 1998 : proceedings (pp. 406-411). (Lecture notes in artificial intelligence; Vol. 1398). Berlin: Springer. https://doi.org/10.1007/BFb0026711