Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!know!zaphod.mps.ohio-state.edu!usc!ucla-cs!math.ucla.edu!barry@pico.math.ucla.edu
From: barry@pico.math.ucla.edu (Barry Merriman)
Newsgroups: comp.ai.neural-nets
Subject: Understanding NN operation
Message-ID: <173@kaos.MATH.UCLA.EDU>
Date: 26 Jul 90 23:44:33 GMT
Sender: news@MATH.UCLA.EDU
Distribution: na
Organization: UCLA Department of Math
Lines: 11


What do you think is the most productive way of 
thinking about how Neural Nets work (for, say, pattern recognition 
or classification tasks)?

By productive, I mean useful for designing or modifying nets, or
for guessing how well a neural net approach will work for a problem.


Thanks,
Barry Merriman