Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!swrinde!pipex!news.sprintlink.net!news.indirect.com!slip39.indirect.com!user
From: hal@indirect.com (Hal Hamilton)
Subject: HELP-Memory Efficiency with CAM
Message-ID: <hal-010595210920@slip39.indirect.com>
Sender: usenet@indirect.com (Internet Direct Admin)
Organization: Internet Direct, indirect.com
Date: Tue, 2 May 1995 04:19:46 GMT
Followup-To: comp.ai.neural-nets
Lines: 14

Can anyone refer me to articles or information on any fundamental
limitations on memory efficiency with artificial neural nets, either as
content addressable memory or in other applications? One of my colleagues
believes that she recalls a limitation of about 40% for Hopfield networks.
This may be a clue about a different kind of problem we are addressing. We
are wondering if this is related to losses in orthoganolizing the data, or
is it more generally or fundamentally related to distributed memory?

Thanks in advance for any help you can provide...

----------------
Hal Hamilton
hal@indirect.com
  
