Capacity of the upstart algorithm

West, Ansgar H L and Saad, David (1997). Capacity of the upstart algorithm. IN: Mathematics of Neural Networks: Models, Algorithms and Applications. Ellacott, Stephen W.; Mason, John C. and Anderson, Iain J. (eds) Oxford: Kluwer.

Abstract

The storage capacity of multilayer networks with overlapping receptive fields is investigated for a constructive algorithm within a one-step replica symmetry breaking (RSB) treatment. We find that the storage capacity increases logarithmically with the number of hidden units <span class='mathrm'>K</span> without saturating the Mitchison-Durbin bound. The slope of the logarithmic increase decays exponentionally with the stability with which the patterns have been stored.

Publication DOI: https://doi.org/10.1007/978-1-4615-6099-9_65
Divisions: Engineering & Applied Sciences > Mathematics
Engineering & Applied Sciences > Systems analytics research institute (SARI)
Additional Information: The original publication is available at www.springerlink.com
Uncontrolled Keywords: algorithms,design,experimentation,human factors,measurement,performance,reliability,security,theory,werification
Full Text Link:
Related URLs: https://link.sp ... -4615-6099-9_65 (Publisher URL)
Published Date: 1997
Authors: West, Ansgar H L
Saad, David ( 0000-0001-9821-2623)

Download

[img]

Version: Accepted Version


Export / Share Citation


Statistics

Additional statistics for this record