Newsgroups: comp.ai.neural-nets,sci.stat.math
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!solaris.cc.vt.edu!newsfeed.internetmci.com!swrinde!cs.utexas.edu!utnut!utgpu!utstat!orourke
From: orourke@utstat.toronto.edu (Keith O'Rourke)
Subject: Re: The book I want to buy
Message-ID: <DJuAqM.MG8@utstat.toronto.edu>
Organization: U of Toronto Statistics
References: <4ar982$612@lace.colorado.edu> <4as159$jp7@newsbf02.news.aol.com> <4avoie$dnt@ixnews4.ix.netcom.com>
Date: Tue, 19 Dec 1995 15:23:10 GMT
Lines: 40
Xref: glinda.oz.cs.cmu.edu comp.ai.neural-nets:28741 sci.stat.math:8433

<A statistician would find a ready market for an accessable book
<about the statistical underpinnings of the models we programmers call
<neural networks. It doesn't have to be easy to understand -- that may
<not be possible -- 

Simple and accessable are not the same thing (as you point out)

Let me relate some personal experience about this ... 

I started out trying to do something like this using a popular 
symbolic algebra package as an excuse to try (also thinking 
that it might appeal to computer science buffs interested in 
neural nets)  

What I ended up trying to do was to make accessable notebooks on 
set theory, functions (from sets to {0,1} i.e. measure theory),
various representations of functions (distributons & moment 
generating functions) and convergence (esp. convergence in 
distribution) ... 

Stuff I am not that qualified in BUT MORE importantly stuff others 
are very unlikely to believe could ever be important until they 
already have a fundamental grasp of statistics.  

The key to statistics, I believe, is the realization that it is 
built upon an MATHEMATICAL model of UNCERTAINTY (here one has to 
grasp both the mathematics and the uncertainty that it models)

Now there is a long history of attempts to make mathematics more 
accessable (along with a few thousand failed attempts).

But to end on a positive note, it was the inventor of the filament 
for light bulbs that lasted more than a few minutes who said afterwords
"if I had understood the difficulty in what I was attempting to do 
I would never have tried"

Keith O'Rourke
The Toronto Hosp.
(Really me no matter whose name appears in the header due to a system bug)

