Title
Performance evaluation of variations to the standard back-propagation algorithm
Abstract
A number of techniques have been proposed recently, which attempt to improve the generalization capabilities of Back-propagation neural networks (BPNNs). Among them, weight-decay, cross-validation, and weight-smoothing are probably the most simple and the most frequently used. This paper presents an empirical performance comparison among the above approaches using two real world databases. In addition, in order to further improve generalization, a combination of all the above approaches has been considered and tested. Experimental results illustrate that the coupling of all the three approaches together, significantly outperforms each other individual approach.
Publication Date
12-1-1994
Publication Title
Southcon Conference Record
Number of Pages
71-76
Document Type
Article; Proceedings Paper
Identifier
scopus
Personal Identifier
scopus
Copyright Status
Unknown
Socpus ID
0028737033 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/0028737033
STARS Citation
Karkhanis, Parag and Bebis, George, "Performance evaluation of variations to the standard back-propagation algorithm" (1994). Scopus Export 1990s. 48.
https://stars.library.ucf.edu/scopus1990/48