Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech2!news.mathworks.com!uunet!in2.uu.net!halon!sybase!usenet
From: George Van Treeck <treeck@sybase.com>
Subject: Re: NNs implemented in PROLOG?
Content-Type: text/plain; charset=us-ascii
Message-ID: <30C5E292.4FA9@sybase.com>
Sender: usenet@sybase.com
Content-Transfer-Encoding: 7bit
Organization: Sybase, Inc.
References: <mreddy-2811951033230001@mreddy.comp.glam.ac.uk> <817590170.21393@roja.demon.co.uk> <mreddy-0112951337420001@mreddy.comp.glam.ac.uk> <49qcrs$g53@cantaloupe.srv.cs.cmu.edu>
Mime-Version: 1.0
Date: Wed, 6 Dec 1995 18:36:02 GMT
X-Mailer: Mozilla 2.0b2 (X11; I; SunOS 5.3 sun4m)
Lines: 8

I know of no NNs implemented in PROLOG, though it should be
fairly simple. Changing the weights would require use of PROLOG's
"retract" and "assert" predicates to implement destructive
assignment. However, NNs are number crunching intensive and
PROLOG is designed for symbolic intensive applications. Performance
of NNs using PROLOG would be terrible.

-George
