From: Eliezer Yudkowsky (sentience@pobox.com)
Date: Mon Jan 27 1997 - 20:57:24 MST
[Omega wonders:]
> I can see why "randomness" might be called "computational temperature"
> but I don't follow the logic that leads this to be equated with human
> emotion, after all, emotion doesn't seem random enough to view this
> way. Could you elaborate on this? Does the book you mention go into
> this?
Computational temperature determines the probability that a
compu-cognitive structure (such as a "bond" or "group") will break under
pressure. In a program called "Jumbo" which is meant to solve anagrams
in a way similar to humans, letters will glom together (such as 't' and
'h' forming "th"). If everything seems to be falling into place, the
computational temperature drops, thus preserving good structures. If
things don't work, the computational temperature rises until the bad
structures (such as "th" in "rathe", the anagram of "heart") break up,
leaving the primitives 'a', 'r', and 't' to form the syllable "art".
The point is that computational temperature is a general, global, sense
which universally affects the actions being taken and to what designed
purpose those actions are directed. Emotion, to coin a phrase.
--
sentience@pobox.com Eliezer S. Yudkowsky
http://tezcat.com/~eliezer/singularity.html
http://tezcat.com/~eliezer/algernon.html
Disclaimer: Unless otherwise specified, I'm not telling you
everything I think I know.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 14:44:05 MST