From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!agate!stanford.edu!rutgers!netnews.upenn.edu!libra.wistar.upenn.edu Thu Dec 26 23:58:30 EST 1991
Article 2398 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!agate!stanford.edu!rutgers!netnews.upenn.edu!libra.wistar.upenn.edu
>From: weemba@libra.wistar.upenn.edu (Matthew P Wiener)
Newsgroups: comp.ai.philosophy
Subject: Chemical neural networks
Message-ID: <61055@netnews.upenn.edu>
Date: 24 Dec 91 23:23:18 GMT
Article-I.D.: netnews.61055
Sender: news@netnews.upenn.edu
Reply-To: weemba@libra.wistar.upenn.edu (Matthew P Wiener)
Organization: The Wistar Institute of Anatomy and Biology
Lines: 25
Nntp-Posting-Host: libra.wistar.upenn.edu

Apropos my previous claim that NN models of olfaction have to show direct
neurobiological implementation in order to be universally believed, and
that even then, they might not generalize to other aspects of cognition,
I came across the following: A Hjelmfelt, E D Weinberger, J Ross "Chemical
implementation of neural networks and Turing machines" PROC NATL ACAD SCI
USA v88, pp 10983-10987, Dec 91.

Briefly, they describe how a cyclic enzyme system known to model the
McCulloch-Pitts neuron (with two steady state concentrations used to
represent the fire/nofire states) can, in theory, be generalized to
model neural networks, whether with continuous or discrete time.  Their
model requires 4 + 4*(# neurons) + (# connections) distinct chemical
species--presumably numerous polymers.  In a spatially distributed
network, the number of distinct species can be reduced immensely.

They claim to have worked out an explicit universal Turing machine in
terms of their model, and will publish the details in a future article.

Presumably fun stuff can be done with neuron+enzyme based neural networks.

It is papers like this that make me dubious of claims that we've almost
deciphered cognition.  Evolution is beholden to no one's theory of mind,
and has had many mental templates to explore.
-- 
-Matthew P Wiener (weemba@libra.wistar.upenn.edu)


