X-Message-Number: 15335
Date: Sun, 14 Jan 2001 11:27:48 -0700
From: Mike Perry <>
Subject: Reply to Thomas Donaldson

Thomas Donaldson, #15328, says:

>First of all, I would consider the THEORETICAL possibility of imitating
>a brain with a single computer to be quite useless unless we could
>somehow make it a real possibility.

Good. I have asked the question before of whether you are only interested in
practical ways of imitating a brain. It appears the answer is yes, and if
so, I don't see much basis for disagreement between our views, assuming a
few other difficulties like the below are resolved, and except insofar as I
am also interested in theoretical quesions too.

...
>Second, the comments about polynomial growth are wrong.
... > the exponentiality comes not from the creation of new neurons,
>but from the number of connections which they allow. ...
>Given a set of N neurons, the number of connections between them goes
>up like N!. 

Here's where I need some clarification. If we have N neurons, it seems to me
that at most we could have every one connected to every other one, and we
could distinguish between a connection going from neuron A to neuron B vs. B
to A. I.e. we would have a complete, directed graph on N vertices. In that
case the number of connections is just N*(N-1) (or N^2 if you allow a neuron
to connect to itself), not N!. It's possible that one "connection" could
involve more than one connector; if that is the case then I suppose any
number of connections could occur between even 2 neurons, let alone N, but
then it seems that a similar effect could be achieved by varying connection
strengths using only one connector. Again, some clarification is needed.

Mike Perry

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=15335