Learning behavior and temporary minima of two-layer neural networks

Annema, Anne-Johan and Hoen, Klaas and Wallinga, Hans (1994) Learning behavior and temporary minima of two-layer neural networks. Neural Networks, 7 (9). pp. 1387-1403. ISSN 0893-6080

Preview

PDF
1798Kb

Abstract:	This paper presents a mathematical analysis of the occurrence of temporary minima during training of a single-output, two-layer neural network, with learning according to the back-propagation algorithm. A new vector decomposition method is introduced, which simplifies the mathematical analysis of learning of neural networks considerably. The analysis shows that temporary minima are inherent to multilayer networks learning. A number of numerical results illustrate the analytical conclusions.
Item Type:	Article
Copyright:	© 1994 Elsevier Science
Faculty:	Electrical Engineering, Mathematics and Computer Science (EEMCS)
Research Group:	Integrated Circuit Design (ICD)
Link to this item:	http://purl.utwente.nl/publications/15176
Official URL:	http://dx.doi.org/10.1016/0893-6080(94)90087-6
Export this item as:	BibTeX EndNote HTML Citation Reference Manager

Show download statistics for this publication

Repository Staff Only: item control page

Metis ID: 112050