Copy Link
Add to Bookmark
Report
Alife Digest Number 019
ALIFE LIST: Artificial Life Research List Number 19 Wednesday, May 9th 1990
ARTIFICIAL LIFE RESEARCH ELECTRONIC MAILING LIST
Maintained by the Indiana University Artificial Life Research Group
Contents:
Steven Strauss article
Boltzmann Tournament Selection and Adaptive Default Hierarchies
Move Over, Darwin
----------------------------------------------------------------------
Date: Fri, 27 Apr 90 17:53:10 EDT
From: RAY <ray@vax1.udel.edu>
Subject: Steven Strauss article
The article on AL by Steven Strauss will appear in the Globe & Mail
(published in Toronto) on Saturday April 28. It will be a few thousand
words long.
Tom Ray
University of Delaware
School of Life & Health Sciences
Newark, Delaware 19716
ray@vax1.acs.udel.edu
302-451-2753
------------------------------
Date: Fri, 04 May 90 09:36:15 CDT
From: Dave Goldberg <DGOLDBER@UA1VM.ua.edu>
Subject: Boltzmann Tournament Selection and Adaptive Default Hierarchies
A technical report entitled "A Note on Boltzmann Tournament Selection
for Genetic Algorithms and Population-oriented Simulated Annealing"
(TCGA Report No. 90003) describes a new form of tournament selection
that achieves a Boltzmann distribution across a population of structures
in a genetic algorithm or in a parallel simulated annealer. The
mechanism requires logistic acceptance and anti-acceptance competitions
for its operation, and proof-of-principle tests and fixed-point theory
show that the mechanism can achieve stable Boltzmann-distributed
populations. The report contains sample Pascal code and suggests the
use of the procedure in parallel and as another form of niching.
A paper presented at the Arizona conference on AI, Simulation and
Planning in High Autonomy Systems describes a mechanism for achieving
accurate default hierarchy formation without reliance on
specificity-based bidding structures. The paper "Reinforcement Learning
with Classifier Systems" by R. E. Smith and myself deescribes the
calculation of a separate priority measure within the classifier that
permits formation (as opposed to discovery) of multi-level DHs that
separate adaptively, thereby permitting near-perfect performance in
the presence of noise, bad rules, or other disturbances.
Copies of these papers are available upon request.
Dave Goldberg
dgoldber@ua1vm.ua.edu or dgoldber@ua1vm.bitnet
------------------------------
Date: Mon, 7 May 90 12:19:34 -0400
From: fritz_dg%ncsd.dnet@gte.com
Subject: Move Over, Darwin
Didn't see the article on AL in Globe and Mail (Digest 17), however, the
Washington Post published one on Sunday, May 6 called "Artificial Life:
Improving on Creation", with the subheading, "Move Over, Darwin: ..." (huh?)!
The article was interesting, and somewhere on the continuum from
favorable to sensational(izing). It was written by an Outlook section
editor, Curt Suplee, and is an excellent piece of journalism.
It touched on most of the major points, and illustrated them well.
However, I question whether the AL community is best served by articles
of this type.
The article quotes (a respected scientist at a major US university)
that by the year 2000 "we're going to see robot beings infiltrate
our everyday lives-- just as microprocessors have come into the toy
industry... Eventually you'll have this colony living in your house,
just sitting there ready to do stuff for you."
Well, that initial date is a short 10 years off, and there are other
things which must come about, such as manufacturing (incubation) techniques,
mass (re)production, market (ecological niche) penetration and
advertising (interspecies signalling), etc., etc.
For years, the National Lampoon made good money parodying these kinds of
exhuberant projections, using Life magazine articles from the 30's, 40's,
and 50's as raw material.
I'll mention two other points. One is the tendency in public forums to
set up straw men, really big and imposing-looking straw men, for a little
heroics. Whence the "move over, Darwin" headline, supported by (academic
community) quotes within the article. The other point is the potential danger
of "eating the heart" out of the AL field.
With enough popularization, the AL topic will come to the attention of the
major dieties of American science, the funding Program Managers. Rapidly,
every second R&D proposal fielded by the nation's scientists and engineers
will have artificial life somewhere in the title. Something like what has
happened to the Neural Network field, in a scant 24 months, will then
occur, where there are suddenly more annual conferences than anyone could
possibly attend, publishers have initiated more NN journals than anyone
with work to do could take the time to track, and most currently funded R&D
efforts in neural networks have nothing to do with neurons, and bear little
resemblance to networks.
My education and post-graduate training was as a biologist and evolutionary
theorist. I've made a career of computational AI. Having watched the
coalescing of various lines of work and thought into AL has touched on ideas
that have been dear to my heart for a long time. I'd hate to see the field
get out of hand, and the name AL become paper currency.
IMHO, stuff where it takes a while to figure out the fundamentals, unlike
something like the Apollo Program where Goddard had done the fundamentals
decades earlier, is not well served by the crash program approach. It's not
certain that a good definition of AL even exists (eg. Digest 17). It may
also be that America's approach to boom and bust science, versus quiet
steady progress, is no longer viable in today's world.
I don't own shares in the corporation AL, but if I did, my vote would be for
a little more substance first before the bandwagon gets underway.
Dave Fritz fritz_dg%ncsd@gte.com
Washington, D.C.
------------------------------
End of ALife Digest
********************************
