From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!yale.edu!spool.mu.edu!hri.com!noc.near.net!garbo.ucc.umass.edu!dime!chelm.cs.umass.edu!yodaiken Sun Dec  1 13:05:27 EST 1991
Article 1630 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca sci.philosophy.tech:1147 comp.ai.philosophy:1630
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!yale.edu!spool.mu.edu!hri.com!noc.near.net!garbo.ucc.umass.edu!dime!chelm.cs.umass.edu!yodaiken
>From: yodaiken@chelm.cs.umass.edu (victor yodaiken)
Newsgroups: sci.philosophy.tech,comp.ai.philosophy
Subject: Re: Zeleny's argument DOA (Was Re: Daniel Dennett)
Message-ID: <39822@dime.cs.umass.edu>
Date: 26 Nov 91 16:16:04 GMT
Article-I.D.: dime.39822
References: <DAVIS.91Nov24033509@passy.ilog.fr> <1991Nov24.124945.5834@husc3.harvard.edu> <1991Nov26.004802.1394@hilbert.cyprs.rain.com>
Sender: news@dime.cs.umass.edu
Followup-To: sci.philosophy.tech
Organization: University of Massachusetts, Amherst
Lines: 18

In article <1991Nov26.004802.1394@hilbert.cyprs.rain.com> max@hilbert.cyprs.rain.com (Max Webb) writes:
>In article <1991Nov24.124945.5834@husc3.harvard.edu> zeleny@zariski.harvard.edu (Mikhail Zeleny) writes:
>>[Assume we can build a machine that exhibits our behavior and competence].
>>However, at any given time, by assessing its construction, we may comprehend
>>all causal factors that influence its behavior (to the extent that this is a
>>machine constructed by ourselves, I assume that we can do so, retracing, if
>>necessary, the modifications imposed on the initial configuration by the
>>learning process).
>
>You assume wrongly. Even now, the behavior of only moderately complex
>ANNs is often impossible to understand by looking only at the modified
>weights (or even the history of those weight changes). I strongly suspect
>that the more sophisticated and complex ANNs to come will be even harder
>to analyse by looking at the encodings. Your argument dies right here.
>

If it can be simulated by a digital computer, then it can be translated
into a lookup table. So where does this difficulty come from?


