From: Dani Eder (danielravennest@yahoo.com)
Date: Mon Sep 15 2003 - 08:24:54 MDT
It seems likely to me that a sufficiently accurate
model of the human brain would be capable of AI.
The unknowns of this approach include:
- Is direct modeling of the human brain the most
efficient approach in terms of computational power
required and training time?
- Just how accurate a model is required?
- How many neurons (or synapses) are required?
The experimental approach (lets build it and see
if it works) is a valid way to explore the unknowns.
At worst they will demonstrate that their particular
design falls short in some respect.
If they get some level of performance out of it -
for example equivalent to maze-running rats -
it will provide a benchmark. A competing approach
can then see if it is getting better or worse
results with a given level of computational power.
This will provide guidance on which approach is
more likely to succeed at the human level.
Also, if they get some measurable results, you
could then extrapolate how much farther the
neural simulation approach needs to go.
Daniel
__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:42 MDT