Copy Link
Add to Bookmark
Report
Alife Digest Number 052
ALIFE LIST: Artificial Life Research List Number 52 Tuesday, January 29th 1991
ARTIFICIAL LIFE RESEARCH ELECTRONIC MAILING LIST
Maintained by the Indiana University Artificial Life Research Group
Contents:
from chris langton
al questions / request for info
A PhD in Alife in the UK?
Two Technical Reports Available from IlliGAL
Call for papers 1st Worshop Principia Cybernetica Project
Darwinian Metaphysics (;-) etc.
Public-domain Simulator Request
----------------------------------------------------------------------
Date: Sun, 6 Jan 91 10:30:49 -0500
From: Marek Lugowski <aliferg@iuvax.cs.indiana.edu>
Subject: from chris langton
At long last, here is the list of the papers that will be published in the
proceedings of the second workshop on Artificial Life, held in Santa Fe,
New Mexico, Feb. 5-9, 1990. A copy of this announcement has been sent to all
contributors. The proceedings themselves should be available sometime in March
or April of 1991.
We apologize for the long delay in selecting the final set of papers. The
review/revision cycle took *much* longer than anticipated.
As was the case with the first proceedings, we received many more papers than
we could possibly publish. Over 60 papers were submitted, about 40 of which
made it through the review process relatively intact. We would love to have
published all 40 of these papers, but this was simply not possible. Therefore,
overall diversity and breadth of coverage were major factors in the selection
of the final set of papers. As with the first proceedings, we wanted these
proceedings to reflect the wide range of approaches that were displayed at the
workshop itself. We also gave preference to work which was NOT reported in the
proceedings of the first workshop, unless there had been significant new
progress to report.
Please keep in mind that a journal on Artificial Life will be starting up
within the next year, and we hope that you will consider submitting future work
to that journal. Also, the 3rd Artificial Life workshop will be held in the
Spring of 1992. We will be sending more information on the Alife journal and
the next workshop early next year, when more details have been arranged.
Finally, our thanks to everybody for all the time and effort that you have put
into your contributions to these proceedings. We are very happy with the overall
scientific quality of the result, and are confident that this book will help to
establish the field of Artificial Life as a vital area of scientific research.
Thanks again!
Christopher G. Langton
(for the Editors)
--------------------------------------------------------------------------------
Papers selected for the Proceedings
of the
Second Artificial Life Workshop
Interaction between Learning and Evolution
D.H. Ackley and M.L. Littman
Emergence of Robust Autocatalytic Networks
R.J. Bagley and J.D. Farmer
Elements D'Epistemologie Fabulatoire
L. Bec
Evolving Networks: Using the Genetic Algorithm with Connectionist Learning
R.K. Belew, J. McInerney, and N.N. Schraudolph
Selfstructuring and Selection: Spiral Waves as a Substrate for Prebiotic Evolution
M.C. Boerlijst and P. Hogeweg
Emergence and Artificial Life
P. Cariani
AntFarm: Towards Simulated Evolution
R.J. Collins and D. Jefferson
Analysis and Simulation of the Development of Cellular Layers
M. De Boer, D. Fracchia, and P. Prusinkiewicz
Artificial Life: The Coming Evolution.
J.D. Farmer
Agorithmic Chemistry
W. Fontana
Co-Evolving Parasites Improve Simulated Evolution as an Optimization Parameter
W.D. Hillis
Learning in the Cultural Process
E. Hutchins and B. Hazlehurst
The Genesys System: Evolution as a Theme in Artificial Life
D. Jefferson, R. Collins, C. Cooper, M. Dyer, M. Flowers, R. Korf,
C. Taylor, and A. Wang
Coevolution to the Edge of Chaos
S. Kauffman
Genetic Evolution and Co-Evolution of Computer Programs
J.R. Koza
Life at the Edge of Chaos
C.G. Langton
Evolutionary Phenomena in Simple Dynamics
K. Lindgren
Synthetic Ethology: An Approach to the Study of Communication
B. MacLennan
Measurement of Evolutionary Activity and Teleology
M.A. Bedau and N.H. Packard
Dynamics of Programmable Matter
S. Rasmussen, C. Knudsen, and R. Feldberg
An Approach to the Synthesis of Artificial Life
T.S. Ray
Evolutionary Optimization in an Artificial RNA World
P. Schuster
Simple Nontrivial Self-Reproducing Machines
A.R. Smith
Learning From Functionalism
E. Sober
Computer Viruses - A Form of Artificial Life?
E.H. Spafford
"Non-optimality" via Preadaptation in Simple Neural Systems
E.G. Stork, B. Jackson, and S. Walker
Evolution of Communication in Artificial Organisms
G.M. Werner and M.G. Dyer
------------------------------
Date: Fri, 11 Jan 91 10:05:07 EST
From: clairday@eds.com (Matt Clairday)
Subject: al questions / request for info
I have never posted to this group, but I must say first that it is a
fascinating area, and I very much enjoy reading about this field of
research. I have several questions/ideas:
1) Are there any small, experimental programs in C or shell scripts that
demonstrate basic AL principles? I would like to get examples that could
be run on UNIX workstations (apollo, hp, sun). Is anything available on
the net?
2) I read in the group previously about a 1980 NASA report describing a
self-replicating lunar factory. I wonder if anyone can tell me if this
report develops a business case for privately funding such a venture.
It seems to me that a consortium of companies might have an interest in
this. Self-replicating machines offer phenomenal prospects for increasing
productivity and the standard of living for all of us. I would
appreciate any more info on this. Perhaps we could brainstorm a
business proposal to put this together !
3) Are there companies that have al products available now in the marketplace?
I am especially wondering about products that address niche applications.
Please email to me, and I can post a summary, if there is general interest.
Thanks.
Matt Clairday
EDS Corporation
Troy, Michigan
------------------------------
From: Geoffrey Ballinger <aighb@castle.edinburgh.ac.uk>
Subject: A PhD in Alife in the UK?
Date: Sat, 12 Jan 91 19:52:52 BST
I am in the final year of my first degree and aim to start a PhD
either in October 91 or, after a year out, in October 92. Does anybody
out there know of anyone who is working in Alife in a UK university and
might be willing to supervise a PhD student? Thanks,
Geoff.
Geoff Ballinger, JANET: Geoff@Uk.Ac.Ed
Department of Artificial Intelligence, UUCP: ...!mcsun!ukc!Ed.Ac.Uk!Geoff
University of Edinburgh, 80 South Bridge,
Edinburgh, Scotland.
------------------------------
Date: Thu, 17 Jan 91 06:34:40 CST
From: "David E. Goldberg" <GOLDBERG@vmd.cso.uiuc.edu>
Subject: Two Technical Reports Available from IlliGAL
Two technical reports are available from the Illinois Genetic
Algorithms Laboratory. "Real-coded genetic algorithms, virtual alphabets,
and blocking" (IlliGAL Report No. 90001), describes a mechanism of convergence
consistent with the schema theorem for real-coded GAs (RCGAs) and
Evolutionsstrategien. The paper suggests that selection chooses
virtual alphabets dimension by dimension during the early stages of the GA;
these alphabets are ultimately searched through the combined
action of selection and recombination. The paper uses this theory to
predict the existence of problems that are blocked from effective RCGA
search, because the global optimum is surrounded by local optima in a
particular way.
"Construction of high-order deceptive functions using low-order Walsh
coefficients"(IlliGAL Report No.90002) constructs partially and fully deceptive
functions using low-order Walsh coefficients. This result debunks the
folktheorem floating around that high-order deception implies high-order
non-zero Walsh coefficients. The construction leads to a more complete
theory of functions of unitation (functions whose value depends only
on the number of ones and zeroes in the argument). These ideas are used
to help explain some of Tanese's puzzling empirical results with parallel
GAs applied to partial Walsh sums of bounded order.
To receive copies of either or both of these reports please send a
request to goldberg@vmd.cso.uiuc.edu. Include your postal address, as
these reports are available in hard copy only.
Dave Goldberg
Dept. of General Engineering
University of Illinois at Urbana-Champaign
117 Transportation Bldg
104 S. Mathews Avenue
Urbana, IL 61801
(217)333-0897
------------------------------
Date: Tue, 22 Jan 91 17:39:40 +0100
From: Z09302%BBRBFU01.BITNET@cunyvm.cuny.edu
Subject: Call for papers 1st Worshop Principia Cybernetica Project
Preliminary Announcement and Call for Papers
*********************************************************
* *
* 1st WORKSHOP OF THE PRINCIPIA CYBERNETICA PROJECT *
* *
* computer-supported cooperative development *
* of an evolutionary-systemic philosophy *
*********************************************************
Free University of Brussels, Belgium
July 2-5, 1991
Organized by: the Principia Cybernetica Editorial Board
the Transdisciplinary Research Group
---------------------------------------------------------------------
Theme
_____
Principia Cybernetica is an attempt by a group of researchers to
collaboratively build a system of cybernetic philosophy, moving
towards a transdisciplinary unification of the domain of Systems
Theory and Cybernetics. This philosophical system will be developed as
a network, consisting of nodes or concepts, linked by specific types
of semantic relations. The network will be implemented in a hybrid
computer-based environment involving hypermedia, electronic mail, and
electronic publishing, thus providing readers and authors with
flexible access to every part of the system.
Development of this system is seen as a long-term project involving
many participants, communicating and conversing through electronic
media, and supervised by an Editorial Board. While traditional
publication of parts of the network will be made periodically, the
project is seen as necessarily open-ended and developing, a process of
discourse among a community of researchers.
The philosophy to be developed is systemic and evolutionary, em-
phasizing the spontaneous emergence of higher levels of organization
or control through variation and natural selection. It includes: 1) a
metaphysics, based on processes as ontological primitives, 2) an epis-
temology, which understands knowledge as constructed by the subject,
but undergoing selection by the environment; 3) an ethics, with the
continuance of the process of evolution as supreme value.
Workshop Topics
_______________
* Supporting Technology: electronic mail and publishing, computer-
supported cooperative working, groupware, Standard Generalized Markup
Languages, hypertext markup languages, hypermedia, object-oriented
environments, ...
* Semantic and Conversational Systems: knowledge representation
schemes for philosophical systems and arguments, different semantic
categories and relations, knowledge structuring and integration,
dealing with disagreements and contradictions, ...
* Constructive Epistemology: model-building, selection criteria for
models, evolutionary epistemology, metacognitive reasoning, levels of
cognition: associative, rational, metarational, ...
* Evolutionary Ethics: survival and immortality as fundamental
values, freedom and/or integration, self-actualization, individual-
society-ecosystem relations, the next metasystem transition: super-
being versus metabeing ...
* Process Metaphysics: modelling of emergence and metasystem
transitions, actions as ontological primitives, the process of evolu-
tion as generator of complexity, levels of organization ...
We would particularly appreciate contributions that cut across these
different tracks: for example, emergence mechanisms applicable to
evolution and to computer-supported knowledge structuring, or
hyperspace as technology and as substrate for "cyberbeings". The
emphasis is on contributions that integrate or synthesize multiple
domains or issues. They are not limited to these separate topics.
Organization of the Workshop
____________________________
After the organization of a symposium on "Cybernetics and Human
Values" at the 8th World Congress of Systems and Cybernetics (New
York, June 1990), the next official activity of the Principia
Cybernetica project will be a Workshop at the Free University of
Brussels (VUB) in July 1991. The official language is English.
The informal Workshop will allow all researchers interested in
collaborating in the Project to meet and to discuss the main problems.
It will be introduced by a more formal Symposium where the main issues
in developing a systemic philosophy will be expounded, as a basis on
which to continue working. The Symposium will take one day, the
Workshop will take two or three days, depending on the number of
contributions. The attendance to the Workshop will be limited to some
35 participants in order to intensify the interactions; the attendance
to the Symposium is unlimited.
The event will be organized in the tradition of a pleasant, infor-
mal setting and warm social contacts initiated by the conference on
"Self-Steering and Cognition in Complex Systems: toward a new
cybernetics" (proceedings edited by Heylighen et al., Gordon and
Breach, New York, 1990), which was held at the same place in 1987.
There will be a possibility for inexpensive accomodation in university
rooms. In addition to that Brussels offers plenty of hotels of all
standards. Interested people may combine participation in the workshop
with participation in the congress of the Int. Fuzzy Systems Assoc.,
which is held at the same location, July 7-12.
Brussels, the capital of Europe, is very easy to reach by a variety
of means of transportation. As the second international city in the
world (measured by the number of headquarters of international
organizations), and with its 1000 years of history, it offers many
interesting sights to visit. It boasts the most beautiful historic
market place and the highest concentration of restaurants in the
world.
Submission of Papers
____________________
Abstracts of minimum one, maximum two pages (about 300-600 words)
should be submitted to one of the addresses below, as soon as possible
but not later than March 15, 1991. If possible, abstracts should be
submitted in printed and in electronic form: email or 3.5 inch floppy
disk (720 Kb MS-DOS or 800Kb Mac) with ASCII text. Authors should give
their full address, phone number, and electronic mail address where
possible. Abstracts will be evaluated by the scientific committee.
Authors will be notified about acceptance before April 15, 1991. It is
possible to send in abstracts late (after March 15), but they will be
considered depending upon whether there are still places available at
the Workshop.
Accepted abstracts will be published in an abstract book available
at the conference. The best contributions will be considered for
elaboration into full papers, to be published in a book devoted to the
Principia Cybernetica project and edited by its present editorial
board. An international scientific publisher is being sollicited.
--------------------------------------------------------------------
Conference Chairman:
___________________
Francis Heylighen (Free University of Brussels)
Scientific Committee:
_____________________
Francis Heylighen (Free University of Brussels)
Cliff Joslyn (State University of New York at Binghamton)
Valentin Turchin (City University of New York)
Jean Paul Van Bendegem (Free University of Brussels)
Gordon Pask (London)
Gerard de Zeeuw (University of Amsterdam)
Jean Ramaekers (Int. Assoc. for Cybernetics, Namur)
Local Organizing Committee:
___________________________
An Vranckx, Eric Van Engeland, Alex Housen
--------------------------------------------------------------------
For submissions of abstracts, or further information, contact:
Francis HEYLIGHEN
PESP, V.U.B., Pleinlaan 2
B-1050 Brussels, Belgium.
Tel. +32 - 2 - 641 25 25
Fax +32 - 2 - 641 22 82
Telex 61051 VUBCO B
Email: Z09302@BBRBFU01.bitnet
Cliff JOSLYN
Systems Science, SUNY Binghamton
6 Garfield Ave. # 2
Binghamton NY 13905, USA.
Tel. +1 - 607 - 729 53 48
Email: cjoslyn@bingvaxu.cc.binghamton.edu
------------------------------
Date: Mon, 28 Jan 91 12:50 GMT
From: BARRY MCMULLIN <75008378%dcu.ie@pucc.princeton.edu>
Subject: Darwinian Metaphysics (;-) etc.
Darwinian Metaphysics,
Replicators,
Vehicles,
and
The Evolution of Evolvability.
[Alife Digest has been a bit quiet of late, so here's an attempt to
stimulate some discussion...]
Richard Dawkins introduced the notion of the ``evolution of
evolvability'' in his paper of that name at the first Alife workshop
(1987). In that paper he suggests that, contrary to his own previous
intuitions, some form of *cumulative* species selection may be
significant for the emergence of such ``evolvability''.
I'm trying to sort out what Dawkins really means (or should mean?) in
terms of his own darwinian metaphysics of replicators and vehicles. He
does not comment on this point explicitly in his workshop paper. In the
new edition of ``The Selfish Gene'' (1989) he refers to the workshop
paper as follows (p. 269):
... Of course, the evolving that we are talking about here is still the
same old evolution, mediated via selection on genes ... There can be a
kind of higher-level selection for embryologies that lend themselves to
evolution: a selection in favour of evolvability...
I won't quote the whole paragraph. He does not explicitly refer to
replicators or vehicles here either, but, reading between the lines, he
seems to equivocate between saying evolution of evolvability may
involve a distinct kind of replicator (``species evolution''?), and
saying that it does not (the vehicles may be different, but we're
talking about the same old replicators).
It is in ``The Extended Phenotype'' (1982) that Dawkins provides his
most comprehensive discussion of replicators and vehicles (isn't it?),
and this is also the work he cites in his workshop paper as indicating
his earlier doubts about the possibility of cumulative species
selection. He seems somewhat ambivalent here also. On p.82 he boldly
states that ``groups of organisms [inc. species?] are best not regarded
as replicators; they are *vehicles* in which replicators travel about'';
but, by p.109 we have the concession (if concession it be) that ``there
may be a case for regarding the gene-pool of a reproductively isolated
group, such as a species, as a replicator'' (this is then followed by
the rider that, at that time at least, Dawkins thought that this, if it
happened at all, was still ``unlikely to explain complex adaptation'').
To restate my question: in the context of the argument for the evolution
of evolvability, should we think of the replicators as good old genes
(small genetic fragments), or as gene pools, or as something else, or
does it matter? Dawkins has not been explicit on this, as far
as I can see, so I'd appreciate any opinions or arguments.
[PS: you all may wonder what the point of this rather intricate query
is---and whether it has any relevance to Alife. Briefly, I suggest that
nothing in (artificial) biology makes sense without (artificial)
evolution; therefore we need to abstract evolution from its original
(terrestrial) biological context; to do this we need, among other
things, an abstract metaphysical foundation such as that offered by
Dawkins (perhaps unwittingly); and I'm having *severe* problems
digesting the latter, which problems are manifested in one small way by
the question raised above. Sigh...]
[PPS: at the end of the day, I don't really have that much confidence in
Dawkins' metaphysics as an appropriate vehicle (;-) for use in Alife at
all; but that is a discussion for some other time---like after I've
*finished* writing my dissertation for example...]
Thanks,
Barry McMullin,
Dublin City University.
<McMullinB@DCU.IE>
------------------------------
Date: Tue, 29 Jan 91 21:27:59 EST
From: Roberto Zamparelli <roberto@prodigal.psych.rochester.edu>
Subject: Public-domain Simulator Request
I am interested in doing experiments using simple neural networks to
"animate" food-searching agents in an artificial environment. The
weights in the networks should be changed by selection, on the base of
how successful the agent was in finding food. There would be NO
LEARNING and NO STRUCTURAL MODIFICATIONS in the networks -- only
evolution from random changes in the weights.
The Universe would be a grid of cells, where food appears and
disappears at random (unless it's eaten).
Does anybody know if there are public-domain simulators that can do this
job, or close enough? C, LISP or PASCAL would be fine.
Since the code should be modified and adapted, a clean, reasonably
well-documented code would be critical.
Thanks,
Roberto Zamparelli
Cognitive Science Program,
University of Rochester
reply to:
roberto@psych.rochester.edu
------------------------------
End of ALife Digest
********************************
