Re: virus: Level Three-Belief and Utility.

David Leeper (
Wed, 02 Oct 1996 16:15:10 -0500

David McFadzean,

: >3) One of the most useful situations for writing Evolution-style computer
: >programs is when standard approaches get stuck in local optima. Why?
: >Evolutionary-style programs do not get stuck in local-optima. I refer
: >you to "An Introduction On Genetic Algorithms". It's written by the
: >brilliant Melanie Mitchell, published by MIT Press and is available at
: >the low cost of $30.00.
: Genetic algorithms most certainly do get stuck in local optima for the
: same reasons you mention below in reference to Hillis's work and also
: because evolution requires selection pressure.

As I pointed out in the next paragraph of that posting, there are times
when simulated evolution gets stuck. But, IMHO, this is because it's a
simulation of the real thing. When compared to the real thing, the
simulation suffers from _severe_ limits on time, memory, storage and
processing power. Even with these severe limitations, GAs tend to get
stuck less often than other methods. This is one of the main reasons
for using GAs.

: Some species haven't
: changed in a billion years because there has been no pressure to evolve.

These species _do_ change. Evolution never stops. Mutations occur in the
genes of all living species. Always and in every example.

Take the example of sharks. How many species of shark are there today?
How many will there be a million years from now? If no changes where
occuring there would be only one species of shark. Now and forever.

A niche may exist for billions of years. We may even be able to look at
creatures who inhabited that niche early on and compare them to creatures
who inhabit them later and see a great number of similarities. Perhaps
they're nearly identical. This doesn't mean evolution has stopped. In
fact, I'll bet five bucks that evolution has mutated the descendants of those
early inhabitants and that some of those decendants have moved on to exploit
other niches. In these new niches these descendents continue to mutate in
ways which better exploit their new environment. Meanwhile, in the original
niche, evolution continues. Change continues. If the niche itself remains
unchanged, and early inhabitants were well adapted, then later inhabits will
bear a strong resemblance to the ancestors. In these cases, it's not
evolution that's stuck, but the niche itself. Evolution is still there,
still chugging away, still tampering with new ideas, still producing new

: Last time I checked the real world had a finite number of organisms
: and number of generations.

Yes. But unlike simulations, evolution is not over yet. It continues on,
species after species. Generation after generation. It never stops. I've
never said evolution is infinite (it's not), but it is never-ending.

> BTW, for everyone else who found replies to Mr. Leeper's mail without
> seeing the original, your mail is probably sorted by date. Apparently
> Mr. Leeper is communicating with us from last month.

One of the advantages of being a God. ;->

David Leeper
Homo Deus  
1 + 1 != 2