Newsgroups: comp.ai.neural-nets
From: jimmy@ecowar.demon.co.uk (Jimmy Shadbolt)
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!gatech!swrinde!pipex!demon!ecowar.demon.co.uk!jimmy
Subject: Re: Fast learning algorithm 
Distribution: world
References: <3bmvdk$rkj@nz12.rz.uni-karlsruhe.de>
Organization: Econostat
Reply-To: jimmy@ecowar.demon.co.uk
X-Newsreader: Simple NEWS 1.90 (ka9q DIS 1.21)
Lines: 25
Date: Thu, 8 Dec 1994 10:26:10 +0000
Message-ID: <786882370snz@ecowar.demon.co.uk>
Sender: usenet@demon.co.uk

In article <3bmvdk$rkj@nz12.rz.uni-karlsruhe.de> ig11@fg30.rz.uni-karlsruhe.de writes:

>
>Hi, has anybody yet thought about this:
>The lerning in the human brain is very fast, especially
>concerning visual things like faces or so. And, at least the
>short term memory is nearly error-free. So how does this
>work? What do you think about it?
>Is it a sort of Hopfield-net?
>
>Discussion welcome.
>
>Andreas Wicker.
>
>
        The ultimate form of *fast* learning is the single-shot
        learning where patterns are simply memorised (e.g. in the look-up
        table, CMAC-like). Does similar but more "continuous" (dynamical) 
        algorithms exists? I was told that ART is fast, and similar
        ideas can be found in the phase-locking network community.

        Cheers

        Drago

