Path: utzoo!attcan!uunet!ncrlnk!ncr-sd!hp-sdd!ucsdhub!sdcsvax!beowulf!pluto From: pluto@beowulf.ucsd.edu (Mark E. P. Plutowski) Newsgroups: comp.ai Subject: Re: "Intelligence" Summary: The role of learning Message-ID: <6605@sdcsvax.UCSD.Edu> Date: 8 Jun 89 22:20:34 GMT References: <952@maestro.htsa.aha.nl> <1130@cbnewsh.ATT.COM> <6585@sdcsvax.UCSD.Edu> <960@maestro.htsa.aha.nl> Sender: nobody@sdcsvax.UCSD.Edu Reply-To: pluto@beowulf.UCSD.EDU (Mark E. P. Plutowski) Organization: EE/CS Dept. U.C. San Diego Lines: 36 To my question-> >> To what degree do you agree that learning is >> a necessary condition of intelligence? Frans van Otten writes: >What do you mean with a "neccesary condition" ? Do you mean that >learning is needed for obtaining intelligence, or that intelligence >requires constant learning ? I think that intelligence is in itself >static: it doesn't require learning (remember my much-used example >of the intelligent program). But by learning the intelligent being >can become more intelligent, whatever that might mean. In this >approach, only these question remains (the essential one): what is >the minimum intelligence-configuration ? And: How must this be >achieved ? By learning or by hardwired bootstrap ? >-- All good points. Let me clarify my position: Intelligence is dynamic. A system that is unable to learn from its mistakes is not intelligent. (A system that is able to learn but chooses not to is just plain pig-headed.) A static architecture would be so easy to fool, (once the limits of its knowledge store are ascertained) most people would point to this as solid evidence that it really does not meet that elusive, as-yet-undefined criteria that forms our intuition of what intelligence is. Now, the question becomes: What is the minimal amount of simultaneous knowledge and learning necessary to qualify for "intelligent behaviour?" Both are necessary since an adaptive system could "learn" very quickly without dsplaying intelligent behaviour, for instance, by forgetting simple basics.