Re: Help

From: Adrian Tymes (wingcat@pacbell.net)
Date: Wed Sep 05 2001 - 21:12:31 MDT


pepe cortisona wrote:
> Hello my name is Alexander Ospina Martínez, i am physician and a
> telecommunications engineer, working on artificial neural networks to
> study
> the learning processes and the neuron patterns. Wh y not in a few years,
> to
> be possible to download these patterns to an artificial device.

It's been done for decades. All you need are some electrodes to
measure the current, some artificial device (almost any general use
computer will do these days) to download the measurements onto, and the
knowhow to wire the one into the other (the technical term is "sensor",
and there are numerous tutorials - both on the Web and in print - on
how to build simple ones).

The challenge is, what do you do with the data once you've captured it?
Neural signals are not simple on/off switches; merely mapping signals
to muscle outputs is a hard task in itself.

Or did you mean to download from an artificial neural net onto
something that interacts with human neurons? If so, then there the
problem is that the artificial net's output does not necessarily have
anything to do with the human neurons, at least until it gets some
feedback data from the neurons - at which point it is connected,
logically if not physically.

> Please send me some information to a pagano8@email.com

Stick around and monitor the list, or delve into the archives.
Searching for info is usually much faster when you actively search,
instead of asking people and waiting for answers.



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:10:24 MST