Computational capabilities of multilayer committee machines

Abstract

We obtained an analytical expression for the computational complexity of many layered committee machines with a finite number of hidden layers (L < 8) using the generalization complexity measure introduced by Franco et al (2006) IEEE Trans. Neural Netw. 17 578. Although our result is valid in the large-size limit and for an overlap synaptic matrix that is ultrametric, it provides a useful tool for inferring the appropriate architecture a network must have to reproduce an arbitrary realizable Boolean function.

Publication DOI: https://doi.org/10.1088/1751-8113/43/44/445103
Divisions: College of Engineering & Physical Sciences > Systems analytics research institute (SARI)
Additional Information: © 2010 IOP Publishing Ltd.
Uncontrolled Keywords: computational complexity,layered committee machines,generalization complexity measure,overlap synaptic matrix,Boolean function,Mathematical Physics,Physics and Astronomy(all),Statistical and Nonlinear Physics,Modelling and Simulation,Statistics and Probability
Publication ISSN: 1751-8121
Last Modified: 03 Jan 2024 17:07
Date Deposited: 14 Dec 2011 12:20
Full Text Link: http://iopscien ... 1/43/44/445103/
Related URLs: http://www.scop ... tnerID=8YFLogxK (Scopus URL)
PURE Output Type: Article
Published Date: 2010-11-05
Authors: Neirotti, Juan P. (ORCID Profile 0000-0002-2409-8917)
Franco, L. Alberto

Download

Export / Share Citation


Statistics

Additional statistics for this record