RE: Is generalisation a limit to intelligence?

From: Samantha Atkins (
Date: Sat Dec 02 2000 - 14:15:16 MST

> -----Original Message-----
> From:
> []On Behalf
> Of Ben Goertzel
> Sent: Saturday, December 02, 2000 6:37 AM
> To:
> Subject: RE: Is generalisation a limit to intelligence?
> > To sum it up: Is there some way to combine the fuzzy quality that
> > intelligence
> > relies on with the rigid quality of not making a single mistake? Is
> > generalisation a limit to intelligence?
> I think this is the same "memory limits intelligence" idea we've been
> tossing
> around, recently?
> With infinite memory, a mind could store both the
> generalization it had made
> from some
> raw data, and all the raw data itself. Thus it could use the
> generalization
> to make predictions,
> and it could also use the raw data to continually re-evaluate its
> generalization to be sure it
> still worked.

That is rather pointless. The entire point of learning, of finding the
explanative pattern was to have something more concise and powerful than
carrying around all the raw perceptions. To really "re-evaluate" would
require more than this anyway. It would require a full in depth simulation
of the entire relevant context. Clearly far too costly. The question is
like asking whether intelligence is a limit to intelligence.

> However, we lack a quantitative science that can tell us
> exactly how quickly
> the error rate approaches
> zero as the memory (&, in a real-time situation, processing
> power able to
> exploit this memory)
> approaches infinity. Eliezer and I differ in that I believe
> such a science
> will someday exist ;>
> We also differ in that he intuits this error rate approaches
> zero faster
> than I intuit it does.

A zero error rate is an impossibility. The universe is not built that way.
The trick is work within these confines successfully. Processing power
approaching infinity (not totally sure what is meant here) is not wholly
relevant. The devil is in all the higher order infinities resulting from
the interactions of all the elements in the infinite memory (universe)
model. Tell me how you propose to account for all of that and have
meta-cycles left for self/model meta-evaluation and evaluation of self-model
interference and interaction and I might begin to believe this idea goes

- samantha

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:35 MDT