Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!gatech!pirates!cssun.mathcs.emory.edu!swrinde!howland.reston.ans.net!news.nic.surfnet.nl!hhinsi!news
From: v922487@si.hhs.nl (Lovi)
Subject: unsupervised learning
Message-ID: <DKx7Jt.I4p@si.hhs.nl>
Sender: news@si.hhs.nl
Reply-To: v922487@si.hhs.nl
Organization: Sector Informatica, Haagse HogeSchool, the NetherLands
Date: Tue, 9 Jan 1996 15:40:41 GMT
Lines: 16


Hello,

Can someone give me more information about the teaching rule of Sanger used in an unsupervised net? I have so far a source code where this rule is implemented, but I want to know the underlying thought of this calculation rule and it's benefits.

Wave,
	Lovi
---
   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   Real name: Remko Riethoven        o/x          e-mail : v922487@si.hhs.nl
   MUDLife: Lovidamo, Piney, Flippy   \  c\_/~~ Lot's of coffee required ;-)
            Jade, GoldDust            /\              Magic: She was my car!
   HypnoDrome Software Development (c) [HDSD]       www.hhs.nl/~v922487/Lovi
   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


