Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!gatech!arclight.uoregon.edu!news.sprintlink.net!news-peer.sprintlink.net!interpath!news.interpath.net!news.interpath.net!sas!newshost.unx.sas.com!saswss
From: saswss@hotellng.unx.sas.com (Warren Sarle)
Subject: Re: Question: probabilistic NNs
Originator: saswss@hotellng.unx.sas.com
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <E4uFL4.3Lt@unx.sas.com>
Date: Thu, 30 Jan 1997 22:53:28 GMT
X-Nntp-Posting-Host: hotellng.unx.sas.com
References:  <5cjjts$4fg@hercules.dic.uchile.cl>
Organization: SAS Institute Inc.
Lines: 20


In article <5cjjts$4fg@hercules.dic.uchile.cl>, nflores@abulafia.ciencias.uchile.cl (Nico Flores) writes:
|> ...
|>    I hope this is not too confusing. In short, I want to learn how to choose
|>    a kernel in order to estimate a given class density. I know this is the
|>    subject of current research, but I would like some basic recipes. In
|>    particular, if I assume a Gaussean kernel with a matrix
|>    proportional to the class covariance matrix, how do I find the optimum
|>    proportionality constant? Notice that this is a one-class problem.

See references under "What is PNN?" in the Neural Network FAQ, part 2 of
7: Learning, at ftp://ftp.sas.com/pub/neural/FAQ2.html

-- 

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.
 *** Do not send me unsolicited commercial or political email! ***

