home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!destroyer!gatech!hubcap!mars!bdbryan
- From: bdbryan@eng.clemson.edu (Ben Bryant)
- Subject: Re: Thanks Neural Nets, and Brains?
- Message-ID: <1992Jul23.222259.9713@hubcap.clemson.edu>
- Sender: news@hubcap.clemson.edu (news)
- Reply-To: bdbryan@eng.clemson.edu
- Organization: College of Engineering, Clemson Univ.
- References: <arms.711907358@spedden>
- Date: Thu, 23 Jul 1992 22:22:59 GMT
- Lines: 49
-
- In article 711907358@spedden, arms@cs.UAlberta.CA (Bill Armstrong) writes:
- >bdbryan@eng.clemson.edu (Ben Bryant) writes:
- >
- >>In article 57397@cc.usu.edu, sl859@cc.usu.edu () writes:
- >
- >>> I do have an additional question. How extensive has the research
- >>>been on neurons of a neural net and the actual neurons of the brain? I
- >>>am interested mostly out of curiosity, but from what I've read I can only
- >>>deduct that we still haven't got a very good model of the actual workings
- >>>of the brain neurons, let alone how to model them with computers.
- >
- >> ... one scientist (I can't remember his name) said that he would
- >> one day like to be able to follow the path of firing responses through
- >>the brain using some MAJOR HEAVY DUTY processing and some HYPER-SENSITIVE
- >>electrical sensors. -I believe when that time comes, this the field of
- >>artificial life (Data from Star Trek?) will appear. That was an article in
- >>one of the spring `92 issues of PS.
- >
- >First off, isn't it rather strange that the most widespread artificial
- >model of neural operation. the multilayer perceptron, uses continuous
- >quantities on its connections, while the dendrites and axons of
- >neurons use "zero or one" type action potentials? Until physiological
- >psychologists start studying adaptive logic networks, can anyone
- >expect much progress on understanding the brain?
-
- Yes, it is very strange that the most widespread artificial model of
- neural operation uses continuous quantities, and I agree whole-heartedly,
- that more research is needed in the way neuronal firings, (mean-rate),
- encode information in the brain. It would seem that these firing
- frequencies (i.e. spikes/sec or some similar unit) actually encode
- what are our thoughts. It will indeed be a major breakthrough when
- we are actually able to trace these firing rates through living brain
- tissue.
-
- >What use the the MAJOR HEAVY DUTY processing if there is no model to
- >help understand the operations? Imagine trying to understand an
- >operating system by looking at the flows of bits in a computer system.
-
- Again, I agree whole-heartedly.
-
- >***************************************************
- >Prof. William W. Armstrong, Computing Science Dept.
- >University of Alberta; Edmonton, Alberta, Canada T6G 2H1
- >arms@cs.ualberta.ca Tel(403)492 2374 FAX 492 1071
-
- Benjamin Bryant
- <bdbryan@eng.clemson.edu>
-
-
-