Re: [sl4] An attempt at empathic AI

From: Vladimir Nesov (robotact@gmail.com)
Date: Mon Feb 23 2009 - 13:14:29 MST


On Mon, Feb 23, 2009 at 4:56 PM, Matt Mahoney <matmahoney@yahoo.com> wrote:
>
> It is possible for an algorithm with high complexity (i.e. there is no simple
> description of it) to do something simple, meaning there is another small
> program that does the same thing. My claim is that an intelligent algorithm
> can't be reduced in such a way. The simplest equivalent program is still complex.
>

In this case there may be no simple algorithm that does the same
thing, there is no small program that does the same thing, that
answers "True" for exactly the same set of inputs. And yet we know
something simple and absolutely true about this complex irreducible
computation. An already running intelligent dynamic, a "Jupiter brain"
for short, if not in information-theoretic sense than at least in
practice may be insanely complex and unpredictable in the specific
things it does. And yet we may be able to know that at each step of
the way it'll hold the property of striving to do the right thing, or
never doing the "Hello world" thing.

-- 
Vladimir Nesov
http://causalityrelay.wordpress.com/


This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:04 MDT