Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Dead Wrong!

Author: Alvaro Polo

Date: 23:16:41 07/22/00

Go up one level in this thread


On July 22, 2000 at 17:32:38, Ralf Elvsén wrote:

>On July 22, 2000 at 16:47:58, Alvaro Polo wrote:
>
>>>Could you please define "complicated".
>>>
>>>Ralf
>>
>>I will do it. The complexity of a system is proportional to the length of the
>>description of its regularities. I don't believe that such description is
>>shorter for any program that for Kasparov's evaluation.
>>
>>Alvaro
>
>Hmm, this sounds interesting. Is this from some theory on computation
>or some such? I don't know much about this. Can this measure be
>applied to physical systems? Can you be just a *little*
>more explicit on what a regularity is?
>
>Oh, I forgot the usual question: any websites? :)
>
>Ralf

I haven't invented this definition. I took it from Murray Gell-Mann (the Nobel
Prize physicist) book "The Quark and the Jaguar", where he states the Theory of
Complexity developed by him and the Santa Fe institute. I read the book in
spanish but I am quite sure that you have it available in english since
Gell-Mann is american.

The complexity measure can be applied to any system that provides information to
us, physical or otherwise.

The subject is long and interesting (in fact I recommend you that if you are
interested you get the book), but I will give a small example, small, on the one
hand, because english is not my langage and a longer explanation could obscure
things, and, on the other hand, because there are a lot of things to say and
this is not the best place to write a 50k message.

Suppose a very "simple" system, a bit string.

For example: 01010101010101010101... (and so on)

This string is not very complex. You can describe its regularities: "a string
where 0 and 1 alternates", a short description.

Now, an even simpler string.

000000000000000000...

This is really simple, "a string composed by 0s".

Now, something more interesting, a random string.

01010010110010011001101010010010... (lets suppose it is aleatory)

Here we could think that it is very complex, but the informational content is
what interest us. An aleatory string has no information and hence it is
equivalent to any other aleatory string. So the description of its regularities
would be "an aleatory string". A random string has the same level of complexity
than a string mades by zeros, since they share the same amount of information.

Sorry to have explained things so badly. If you find the subject interesting you
could search by Gell-Mann and complexity theory and Santa Fe Institute and also
get the book.

Alvaro Polo



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.