Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!godot.cc.duq.edu!newsgate.duke.edu!news.mathworks.com!newsfeed.internetmci.com!in1.uu.net!news.interpath.net!sas!newshost.unx.sas.com!saswss
From: saswss@hotellng.unx.sas.com (Warren Sarle)
Subject: Re: nn and stat decisions for med:  Bayes vs neorotic nets
Originator: saswss@hotellng.unx.sas.com
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <DrxFq5.IJA@unx.sas.com>
Date: Fri, 24 May 1996 20:32:29 GMT
X-Nntp-Posting-Host: hotellng.unx.sas.com
References: <Pine.EPX.3.92.960512171428.13920D-100000@powai.cc.iitb.ernet.in> <319E6EBB.41C67EA6@camis.stanford.edu> <4nngcr$g11@scapa.cs.ualberta.ca> <4nqgls$pg6@tuegate.tue.nl> <31A16580.41C67EA6@camis.stanford.edu> <4o25j2$geg@tuegate.tue.nl>
Organization: SAS Institute Inc.
Lines: 24


In article <4o25j2$geg@tuegate.tue.nl>, Johan Haasakker <h.j.haasakker> writes:
|> ...
|> Ok, I agree that the pdf est need not be too precise, but it should take into
|>    an account stochastic dependence between various cues' events.  Thats what
|> i ment by quality and thats what I ment by well done Bayes.
...
|> I have seen many tables of the results (usually 10 randomly selected splits) and
|> none of them were better than my Bayes.

The phrase "well done Bayes" is no more informative than "well done
neural net". Which of numerous Bayesian estimation methods applied
to which of an infinite number of models does the above refer to?

And was "neorotic nets" intended to be "neurotic nets"? If so, I must
add a reference to the FAQ as soon as someone provides me with one.



-- 

Warren S. Sarle       SAS Institute Inc.   The opinions expressed here
saswss@unx.sas.com    SAS Campus Drive     are mine and not necessarily
(919) 677-8000        Cary, NC 27513, USA  those of SAS Institute.
