Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!news.kei.com!news.mathworks.com!gatech2!swrinde!sdd.hp.com!vixen.cso.uiuc.edu!uwm.edu!lll-winken.llnl.gov!simtel!news00.sunet.se!sunic!news99.sunet.se!mn5.swip.net!seunet!news2.swip.net!sw0198!news
From: Tom Sjquist <tomsj@li.icl.se>
Subject: Re: Travelling Salesman Problem
Content-Type: text/plain; charset=us-ascii
Message-ID: <30ACD160.79C1@li.icl.se>
Sender: news@sw0198.icl.se
Content-Transfer-Encoding: 7bit
Organization: ICL
References: <48flji$stv@hermes.louisville.edu>
Mime-Version: 1.0
Date: Fri, 17 Nov 1995 18:16:00 GMT
X-Mailer: Mozilla 2.0b2 (Windows; I; 32bit)
Lines: 24

Hopfield and Tank looked at the Travelling Saleman Problem (using 
sqr(n) nodes to solve n cities) in the August , 1986 number of 
Science pp 625 to 633. I haven't seen the full text (I saw the 
reference in a course from CEI-EUROPE) but I beleive that the 
solution was "good but not always optimal".

/Tom

David Jansing wrote:
> 
> Does anyone know if the Classic Travelling Salesman Problem has 
been looked
> at using a Neural approach?  If so, can you send me the reference? 
 Thanks.
> 
> --
> Dave Jansing - System Adm.                         | This .sig file 
takes
> dave@edjans01.remote.louisville.edu                | 4 D batteries 
(Not
> http://edjans01.remote.louisville.edu/~dave/       | Included).
> 
---------------------------------------------------------------------
----------
