Shane Legg wrote:
> Whose to say what intelligence *really* is.  My approach is simply
> to try to define something that captures with precision aspects of
> the fuzzy concept of intelligence, name it something else, and then
> get on with the job.  If nobody/everybody thinks that what I have
> defined has the same meaning as the natural language concept of
> intelligence it doesn't make any difference.  My word of choice is
> cybernance (technically the word makes sense though things cyber- are
> getting a bit much).  I defined it in terms of algorithmic information
> theory but later discovered that I'd been beaten to it by somebody else so
> never published my ideas.  (If you take the k=1 case in Marcus Hutter's
> "intelligence order relation" that's what I had in mind - he goes on to
> prove all sorts of interesting and powerful results)  So even if you
> don't think that this is really "intelligence" you would still have to
> agree that it's a very powerful concept --- which is good enough for me.

Hutter's concept of "intelligence order relation" basically is the ability
to maximize functions of an environment, via ongoing actions in that
environment.

This is very close to my notion of intelligence as "the ability to achieve
complex goals in a complex environment."  Achieving complex goals implies
maximizing functions, and vice versa.

The only difference between the two is that they suggest different ways of
quantifying the *extent* of intelligence in a given system.  I explore this
differences in an early chapter of the Novamente book...

Anyway, I agree with Shane that this is a useful sort of concept, and I also
agree that it doesn't really matter how much this or that person believes
the concept corresponds with the ambiguous natural language concept of
"intelligence."

Concepts created for the era when humans and animals were the only known
intelligences, need to be revised as the AGI era dawns ;-)

-- Ben


-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/

Reply via email to