Path: utzoo!attcan!uunet!cs.utexas.edu!yale!mintaka!nuhub!nic!hri!sparc9!rolandi
From: rolandi@sparc9.hri.com (Walter Rolandi)
Newsgroups: comp.ai
Subject: Re: Intelligent Rooms
Keywords: Knowledge Acquistion, Evolution
Message-ID: <1990Jun29.121433.4626@hri.com>
Date: 29 Jun 90 12:14:33 GMT
References: <57861@bbn.BBN.COM> <25445@cs.yale.edu> <1990Jun26.140923.22895@cs.umn.edu> <25457@cs.yale.edu> <22928@dartvax.Dartmouth.EDU>
Sender: news@hri.com
Reply-To: rolandi@sparc9.hri.com (Walter Rolandi)
Organization: Horizon Research
Lines: 30

In article <57861@bbn.BBN.COM>, dredick@bbn.com (Barry Kort) writes:
>
> Indeed.  It doesn't make much sense to talk about an intelligent system whose
> knowledge base is static.  The acquisition of knowledge over time (learning)
> is an important feature of intelligent beings, be they made of silicon or
> made of meat.  The challenge for AI and Cognitive Science is to learn how
> to make the learning process efficient. 
>

This ability is more than an important feature of intelligent beings.
Learning is a process, inferred from the observation of behavioral change,
which as an explanatory construct, *accounts for* intelligent behavior.

AI folks would be better off if they worried less about what knowledge *is*
and more about how a machine might acquire the behaviors from which the 
existence of knowledge is inferred.  Knowledge is what knowledge does.





              ***************************************
              *          Walter G. Rolandi          *
              *        Horizon Research, Inc.       *
              *           1432 Main Street          *
              *       Waltham, MA  02154  USA       *
              *            (617) 466 8339           *
              *                                     *
              *           rolandi@hri.com           *
              ***************************************