From: Krekoski Ross (rosskrekoski@gmail.com)
Date: Sun Jun 01 2008 - 01:46:20 MDT
On Wed, May 28, 2008 at 10:20 AM, Nick Tarleton <nickptar@gmail.com> wrote:
>
> And why on Earth would you define it that way? That's knowledge -
> intelligence is how well that knowledge can be acquired and used.
>
You're making needless distinctions between intelligence and knowledge. My
assertion was simply that in an optimally configured system, regardless of
what algorithms are used (presumably we dont know what algorithms at this
point anyways so its pointless to speculate), one can create a direct
correlation between Kolmogorov complexity and intelligence. The higher the
complexity, the greater the potential intelligence that can emerge from a
system. That said, a given program cannot increase its own complexity
without external input, since the complexity of any program without external
input at any point in time can be described with the original code plus an
integer that describes the number of program run cycles, for example.
The complexity of any given program therefore can be described as a function
of its original complexity, and the additional complexity of any input that
the program received during its runtime. The intelligence of a system
therefore would be subject to precisely the same restrictions, and would
require large amounts of input to increase in intelligence in any meaningful
way.
Ross
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:03 MDT