Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!usc!cs.utexas.edu!solo.csci.unt.edu!ponder.csci.unt.edu!ian
From: ian@ponder.csci.unt.edu (Ian Parberry)
Newsgroups: comp.ai.neural-nets
Subject: reliability in neural networks
Summary: tech report available
Message-ID: <1991Feb20.215150.18888@solo.csci.unt.edu>
Date: 20 Feb 91 21:51:50 GMT
Sender: usenet@solo.csci.unt.edu (Usenet News)
Organization: University of North Texas, Denton
Lines: 29
Originator: ian@ponder.csci.unt.edu

The following technical report is now available from CRPDC:

P. Berman, I. Parberry,  and G. Schnitger, ``A Note on the Complexity of
Reliability in Neural Networks'', Technical Report CRPDC-91-3,
Center for Research in Parallel and Distributed Computing, Dept. of
Computer Sciences, University of North Texas, Feb. 1991.

ABSTRACT: It is shown that in a standard discrete neural network
model with small fan-in, tolerance to random malicious faults can be
achieved with a log-linear increase in the number of neurons and
a constant factor increase in parallel time, provided fan-in can increase
arbitrarily.  A similar result is obtained for a nonstandard but closely
related model with no restriction on fan-in.

Write to:

Technical Reports Librarian
Dept, of Computer Sciences
University of North Texas
P.O. Box 13886
Denton, TX 76203-3886

(Mike Carter, if you are reading this, I seem to have lost your address
in my move to Texas.  If you email it to me at ian@csci.unt.edu, I will
send you a copy as promised).
____
Ian Parberry  ian@dept.csci.unt.edu  Dept. of Computer Science,
Univ. of North Texas, P.O. Box 13886, Denton, TX 76203-3886
"Bureaucracy is expanding to meet the needs of an expanding bureaucracy"