OT: Neural Networks and fp

[Fairlight]

On Wed, Dec 01, 2004 at 09:09:00AM -0500, Walter Vaughan, the prominent pundit,
witicized:
> However, from what I have read, Ken or whoever wrote the code to read 
> the patterns from the fingerprint readers *had* to implement some type 
> of neural network. Now if that code can be leveraged in the future as a 
> function in filePro... well only time will tell.

What makes you think that requires a neural net?  As I understand it,
fingerprints are broken down and analysed by the scan devices themselves
and then translated to a unique code.  That code is simply stored and you
can search for it--you either match or you don't.  I don't see any learning
implied in that scenario.

> On the other hand filepro already acts like a neural analyzer. *report 
> makes two passes if asked, and stores information about the 1st pass as 
> it does it's thing.

That's an algorithm, not actual learning.  

Gnuchess is closer to a nn than this, as it will at least store and use
past games as "experience" if you tell it to.  That's a lot different than
doing something that's simply two passes over the data.

> The framework is there.

I'm kinda doubting that.

mark->
-- 
Bring the web-enabling power of OneGate to -your- filePro applications today!

Try the live filePro-based, OneGate-enabled demo at the following URL:
               http://www2.onnik.com/~fairlite/flfssindex.html