Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!elroy.jpl.nasa.gov!lll-winken!unixhub!shelby!portia.stanford.edu!portia!bellido From: bellido@boromir.world (Ignacio Bellido) Newsgroups: comp.ai.neural-nets Subject: Re: Postulates on the number of neurons Message-ID: Date: 12 Dec 90 03:56:48 GMT References: <1990Dec11.040646.20760@noose.ecn.purdue.edu> <1990Dec11.174921.26838@cs.columbia.edu> Sender: news@portia.Stanford.EDU Reply-To: bellido@psych.stanford.edu Followup-To: comp.ai.neural-nets Organization: /user/bellido/.organization Lines: 53 In-Reply-To: fahn@cs.columbia.edu's message of 11 Dec 90 17:49:21 GMT In article <1990Dec11.174921.26838@cs.columbia.edu>, fahn@hudson.columbia.edu (Paul Fahn) writes: >It seems to me that (computer-based) neural net researchers too often >presume that biological organization is somehow optimal. It may be that >the brain represents an enormous waste of resources, and that >"intelligent behavior" can be achieved with far fewer neurons, if only >they were organized more "intelligently". (or, equivalently, that higher >intelligence could be achieved with the same number of neurons.) > >We should definitely look at biological nets for ideas, but in a very >critical way, and be prepared to reject anything found in the biological >realm. Let's see, I believe that biological organization IS optimal taken acount of it's path, I mean, the selections made along the evolution, the advantage of biological optimization is a lot of million years over us (we just learned some years ago that the brain has cells and that they are independent elements), and a very big number of choices to select and eliminate on this time. May be another choice some millions years ago could have boosted another race to be intelligent with less amount of neurons better organized, but we will never know, and I don't think so. Are there any waste of resources in the human brain? I believe not, Nature trends always to the minimum energy, and it's not the way to waste. Also, in evolution is very important to have a big amount of variety, a really optimal organization probably means a cut of variety and this means less choices. It also means a bad tolerance to failures. There are also a lot of parameters to put into the acount, generality of the basic brain structure, fault tolerance (really important), information capacity, real time learnning, generalization,... These kind of things are very expensive in resources, but they are neccesary for an "intelligent" behavior, so it's not waste. Can we design a better organization? I don't think so, but let see how these field's evolution goes. Ignacio -- -------------------------------------------------------------------------- Ignacio Bellido Fernandez-Montes -1z Visiting Scholar at Stanford University e-mail: bellido@psych.stanford.edu Psychology Department Graduate Student Madrid University of Technology Department of Telematic Engineering e-mail: ibellido@dit.upm.es --------------------------------------------------------------------------