Binmore then goes on to argue that Axelrod engages in
misleading generalization by reiterating his claim, originally made
in The Evolution of Cooperation, that TIT-FOR-TAT embodies the
essential features of a successful strategy for the indefinitely
repeated Prisoners' Dilemma and provides a suitable paradigm for
human co-operation in a much wider range of contexts. Binmore's
criticism takes two basic forms: (a) He argues that several of the
key conclusions that Axelrod draws from simulations turn out not to
be robust to modifications of his experimental design, such as an
increase in the number of iterations or a change in the initial
conditions that puts the multi-agent system within a different basin
of attraction; and (b) he argues that Axelrod has made the mistake,
routinely committed by social scientists who use simulation
techniques, of ignoring the light that game theory per se can shed
on issues such as the evolution of cooperation.
- Axelrod's Reply to Binmore's Review:
- "Ken Binmore reports in his review of The Complexity of
Cooperation that he was irritated by the jacket's claim that I had
done groundbreaking work in game theory. That he was irritated is
clear enough. The basis for his irritation is not clear since later
in the same paragraph he says, `I believe that he did make an
important contribution to game theory [by] focusing our attention on
the importance of evolution in selecting an equilibrium from the
infinitude of possibilities whose existence is demonstrated by the
folk theorem.'
- The book being reviewed is a sequel to The Evolution of
Cooperation. Most of the review actually deals with the first
book where I gave theoretical and empirical reasons to support the
conclusion that reciprocity is a very robust basis for cooperation
among egoists. I invited game theorists from a variety of
disciplines to submit the strategy that they believed would do best
in the environment composed of others who would be submitting such
strategies. Binmore is concerned that in the first round of the
tournament, I told the entrants how many moves each game would have.
From the point of view of traditional game theory where the players
have unlimited rationality, they should defect on the (known) last
move, and therefore on the next to last move, and so on back to the
first move. (I even gave each entrant a written form of this
argument for their consideration.) However, none of the entrants
submitted a strategy that always defects. In fact, they were smart
to ignore standard game theory reasoning since everyone else did.
To the credit of game theory as a discipline, there has been a great
deal of work in recent years (including interesting work by Binmore
himself) on the implications of relaxing the assumption of unlimited
rationality.
- The second specific criticism also applies to The
Evolution of Cooperation. Binmore is not satisfied with the
evidence for the robustness of TIT FOR TAT as an effective strategy
for the iterated Prisoner's Dilemma. I offered the following: two
rounds of the tournament showing TIT FOR TAT succeeded by doing well
with a wide variety of other strategies designed by more or less
sophisticated people, six variants of the tournament emphasizing
different aspects of that environment, and an ecological simulation
that demonstrated that TIT FOR TAT would do well in the long run.
Moreover, the book under review, The Complexity of
Cooperation, offers two more kinds of evidence about the
robustness of reciprocity. Chapter 1 shows that in an evolutionary
setting with mutation and selection, reciprocity can establish
itself from a random start, and Chapter 2 shows that adding a little
generosity or contrition can effectively deal with the problems of
occasional misunderstanding or misimplementation. There has also
been a great deal of empirical research showing that reciprocity is
a common and effective strategy in a very wide variety of settings
including international politics, and among fish, birds, and monkeys
(see R. Axelrod and D. Dion, Science, 9 December 1988).
- Binmore ignores all but the first three of the nine
parts of the book he is reviewing. In the rest, one will find a
concern with empirical reality, including a landscape theory of
alliances that successfully accounts for the alignment of European
nations in World War II, and the alignment of computer companies in
setting standards for UNIX (Chapters 4 and 5). One will also find
simulation models of the formation of new political actors (Chapter
6) and the dissemination of cultures (Chapter 7). Computer
simulation is indeed a fruitful approach for advancing our
understanding of complex social processes."
Other New ACE-Related Books
Note: Cites to the following books can also be found on the
annotated ACE syllabus linked to the ACE Web site home page.
-
H. Peyton Young, Individual and Social Structure: An
Evolutionary Theory of Institutions, Princeton University Press,
Princeton, N.J., 1998, 184 pages, Cloth: 0-691-02684-X.
- From the book blurb: "Traditionally, economists have
assumed an ideal world where people make perfectly rational
decisions based on complete knowledge of their particular situation.
(This book) proposes a new analytical framework based on a more
realistic world in which people have a limited understanding of
their environment, sometimes leading them to do short-sighted or
irrational things. Using game theory, Young postulates that
customs, norms, and forms of organization are built up over long
periods of time from the cumulative experience of many individuals
into recognizable patterns of social behavior. At the macro level,
the theory predicts how institutions will evolve and what
characteristics they will possess. At the micro level, it shows
that individual interactions will often take the form of a game with
multiple equilibria." H. Peyton Young is Scott and Barbara Black
Professor of Economics at Johns Hopkins University and a Visiting
Fellow at the Brookings Institution.
-
J. Hofbauer and K. Sigmund, Evolutionary Games and
Replicator Dynamics, Cambridge University Press, 1998,
0-521-62365-0.
- From the book blurb: "Evolutionary game theory replaces the
static solutions of classical game theory by a dynamical approach
centered not on the concept of rational players but on the
population dynamics of behavioral programs. In this book the
authors investigate the nonlinear dynamics of the self-regulation of
social and economic behavior, and of the closely related
interactions among species in ecological communities."
-
M. Prietula, K. Carley, and L. Gasser, Simulating Organizations:
Computational Models of Institutions and Groups, MIT Press,
Cambridge, MA, 1998, 0-262-66108-X.
- From the book blurb: "Although a great deal of work remains
to be done, the era is approaching when both theorists and
practitioners will routinely state theories, design organizations,
and derive their implications using widely shared computational
tools. This volume brings together a range of work from many of the
leading researchers in the field."
-
R. Dai (ed.), Virtual Reality for Industrial
Applications, Springer-Verlag, Berlin/Heidelberg, 1998, ISGN
3-540-63348-0.
- From the book blurb: "This book gives a systematic overview
of the fundamental aspects of VR technology and its industrial
applications, based on up-to-date information provided by experts
from research and industry.
-
W. T. Anderson, Evolution Isn't What It Used To Be: The Augmented
Animal and the Whole Wired World, W. H. Freeman Press, 233
pages, 1998, 0-7167-3134-7.
- From the book blurb: "Prepare for life in the
rapidly-emerging bio-information society with this fascinating and
well-documented discussion of the information and biological
revolutions that are quietly affecting everything from our genes to
the economic fate of nations."
-
J. A. Anderson and E. Rosenfield (eds.), Talking Nets: An Oral
History of Neurocomputing, MIT Press, 1998, 0-262-01167-0.
- From the book blurb: "Since World War II, a group of
scientists has been attempting to understand the human nervous
system and to build computer systems that emulate the brain's
abilities. ... In this collection of interviews, those who helped to
shape the field share their childhood memories, their influences,
how they became interested neural networks, and what they see as its
future."
-
D. Kortenkamp, R. P. Bonasso, and R. Murphy (eds.), Artificial
Intelligence and Mobile Robots, MIT Press, 1998, 0-262-61137-6.
- From the book blurb: "The mobile robot systems described in
this book were selected from among the best available
implementations by leading universities and research laboratories."
-
S. R. Hameroff, A. W. Kaszniak, and A. C. Scott (eds.), Toward a
Science of Consciousness 2, MIT Press, 1998, 0-262-08262-4.
- From the book blurb: "What is Consciousness? Recent
attempts to answer this question have motivated two
interdisciplinary conferences sponsored by the University of Arizona
in Tucson. ... This volume presents a selection of invited papers
from the second conference, held in April 1996.
-
P. G. Morasso and V. Sanguineti (eds.), Self-Organization,
Computational Maps, and Motor Control, North-Holland, 1997.
- From the book blurb: "In the study of the computational
structure of biological/robotic sensorimotor systems, distributed
models have gained center stage in recent years, with a range of
issues including self-organization, non-linear dynamics, field
computing etc. This multidisciplinary research area is addressed
here by a multidisciplinary team of researchers..."
New ACE-Related Journals
Note: Pointers to the journals listed below can be found
on the journals page linked to the ACE Web site home page.
- Computational and Mathematical Organization Theory (CMOT)
- CMOT provides an international forum of research that
advances organization theory and analysis through the use of
computational and mathematical techniques. The journal presents a
new perspective on organizational research that extends the
traditional mathematical approach to formal organizational theory by
including computer simulation, logic, and artificial intelligence.
- For more information, visit the
CMOT Web site.
- Journal of Memetics: Evolutionary Models of Information
Transmission (JoM-EMIT)
- In 1976, Richard Dawkins invented the word "meme," defining
it as the "new replicator," "a unit of cultural transmission, or a
unit of imitation." JoM-EMIT is a new peer-reviewed academic
journal that seeks to develop the memetic perspective, with space
devoted to relevant evolutionary issues and other related topics.
It offers a forum where theories and the philosophy of memes and
evolution are in the centre, not just at the edge of the issues
journals want to cover. Members of the Advisory Board include Gary
Cziko, Richard Dawkins, Daniel Dennett, and David Hull.
- For more information, visit the
JoM-EMIT Web site.
Note: The next two announcements appeared in the February
1998 newsletter of the Program for the Study of Complex Systems at
the University of Michigan.
- Journal of Complex Systems (JCS):
- JCS aims to provide a medium of communication for
multidisciplinary approaches, either empirical or theoretical, to
the study of complex systems in such diverse fields as biology,
physics, engineering, economics, cognitive science and social
sciences, so as to promote the cross-fertilization of ideas among
all the scientific disciplines having to deal with their own complex
systems. By complex system, it is meant a system comprised of a
(usually large) number of (usually strongly) interacting entities,
processes, or agents, the understanding of which requires the
development, or the use of, new scientific tools, nonlinear models,
out-of-equilibrium descriptions and computer simulations.
Understanding the behavior of the whole from the behavior of its
constituent units is a recurring theme in modern science, and is the
central topic of JCS.
- Papers suitable for publication in JCS should deal with
complex systems, from an empirical or theoretical viewpoint, or
both, in biology, physics, engineering, economics, cognitive science
and social sciences. This list is not exhaustive. Papers should
have a cross-disciplinary approach or perspective, and have a
significant scientific and technical content.
- For further submission information contact Eric Bonabeau,
Editor-in-Chief, at bonabeau@ santafe.edu. Information regarding
subscription should be sent to hermes@iway.fr, attention
subscriptions department.
- Evolutionary Optimization (EO):
- The subject of evolutionary optimization has recently
experienced a remarkable growth. New concepts, methods and
applications are being continually proposed and exploited to provide
efficient tools for solving a variety of optimization problems. The
aim of this international journal is to collect and disseminate the
progressive body of knowledge on evolutionary optimization
techniques and their applications, via a single organized medium.
These techniques include Genetic Algorithms, Genetic Programmming,
Evolutionary Programming and Evolution Strategies among others, all
of which are inspired by restricted models of natural evolution.
- EO will be publishing invited papers, original research
and review papers and short letters. The journal will also have
special issues devoted to relevant topics. Book reviews,
forthcoming events and software sections of the journal will report
the recent developments and advances in the field.
- EO will be published quarterly starting in 1998. For
further information please contact Dr. Sourav Kundu at the Kanazawa
University sourav@kenroku.ipc.kanazawa-u.ac.jp
Conference Announcements
Note: A list of pointers to upcoming conferences can be
found on the conference page linked to the ACE Web site home page.
- Excerpts from a Workshop Summary by Nigel Gilbert
(Department of Sociology, University of Surrey, Guildford, UK)
- The Workshop on the Potential of Computer Simulation for
the Social Sciences, with forty six participants, was held
January 14-15, 1998, at the University of Surrey by the Centre for
Research on Simulation in the Social Sciences (CRESS), sponsored by
the UK Economic and Social Research Council. The aim of the
workshop was to review the opportunities for simulation-based
research and, especially, to publicise and encourage the development
of such research within the UK.
- Workshop speakers included Klaus Troitzsch (origins,
prospects, and purposes of social science simulation), Scott Moss
(computational environments for addressing alternative policy
strategies), Kathleen Carley (analysis of organizations as complex
and adaptive), Rainer Hegselmann (studying social processes by
cellular automata), Rosaria Conte (the PART-NET system for modelling
agents with internal cognitive structure), and Richard Eiser (a
connectionist theory of attitudes and social influence), among
others. A more detailed report on this workshop can be found at the
workshop Web site.
Additional ACE-Related Web Sites
Note: Pointers to the following Web sites are included on the
"other ACE-related Web sites" page linked to the ACE Web site home
page.
- The Program for the Study of Complex Systems at the University
of Michigan provides continually updated information about
postdoctoral openings and upcoming conferences
related to the study of complex systems.
- The
Centre for Economic Learning and Social Evolution (ELSE),
directed by Ken Binmore, is an interdisciplinary research center
based at University College London that promotes the study of models
of interactive learning with the view of providing a new foundation
for modelling behaviour in economics and related social sciences.
Researchers at ELSE have a special interest in game theory and its
applications.
-
The International Society of Dynamic Games,
founded in Helsinki, Finland, in 1990, seeks to promote interactions
among researchers interested in the application of dynamic games.
To this end the society sponsors conferences and workshops, a
newsletter, and an archival journal, Annals of Dynamic Games.
- The
Social Interaction Economics and Computing Web site
is maintained by a group of researchers at the Department of
Economics, University of Ancona, Italy. The focus of the group is
on macroeconomic research that postulates heterogeneous interacting
agents, thus avoiding the use of constructs such as a single
representative consumer or firm. [Note: This is a repeat
announcement with a revised Web address.]
- The
IKE Group
is part of the Department of Business Studies, Aalborg University,
Denmark. The group does research on economic, technical, and
institutional change. The main resarch themes include economic
evolutionary modelling, the theory of the firm, national systems of
innovation, international trade and competitiveness, and the
interplay between economic and ecological issues.
- Researchers at the
Oregon Graduate Institute of Science and Technology
in Portland, Oregon, are exploring the flow of communication and
information into, within, and out of business enterprises, viewed as
complex adaptive systems.
- The
Distributed Artificial Intelligence Research Unit
at Queen Mary and Westfield College (University of London) has
developed and applied agent and multi-agent techniques to real world
problems in a wide range of commercial and industrial domains.
Applications have included: telecommunications network management;
business process management; electricity management; patient care;
concurrent engineering; 3-D scientific data interpretation; digital
libraries; and process control.
- Researchers participating in the
Agents Technology Group
at Hewlett Packard Laboratories are exploring the use of software
agents in electron commerce. As preliminary steps, they are
developing their own agent toolkit called Woad and working on a
formal analysis of the efficiency of various mult-agent systems
architectures.
-
Complexity-On-Line,
an extensive archive of resources on complex systems maintained by
researchers at the Charles Sturt University in Australia.
-
The New England Complex Systems Institute (NECSI),
established in 1997 by a group of faculty in the New England area, is an
independent educational and research institution dedicated to advancing the
study of complex systems.
- The
Center for Connected Learning and Computer-Based Modelling
at Tufts University, Medford, Massachusetts, is a four to five year project
whose goal is to create tools that will help learners (at all levels) to make
greater sense of complex phenomena, and to study how learners come to
understand complexity. In particular, a concrete goal of this project is to
develop "object-based parallel modeling languages" (OBPML), sometimes
referred to as agent-based modeling languages, that can be used by learners
to create rich and detailed models of large systems of interacting agents and
objects. The center has recently released as freeware a new version of the
OBPML StarLogoT (available only for the Macintosh). The original version of
StarLogo was developed under Mitchel Resnik, currently a professor in the
Epistemology and Learning Group (ELG)
at the MIT Media Laboratory. The ELG specializes in the development of new
computational tools that help people learn new things in new ways, e.g.,
after-school learning centers, toys, museum exhibits, and on-line virtual
worlds for exploring and learning about decentralized systems and emergent
phenomena. StarLogoT was developed at Tufts in affiliation with the ELG.
- The
Center for Complex Systems Research (CCSR)
is an interdisciplinary group of faculty and graduate students investigating
a variety of complex dynamic processes that occur in the areas of biology,
physics, chemistry, and astonomy. The CCSR is located at the Beckman
Institute for Advanced Science and Technology at the University of Illinois
at Urbana-Champaign.
-
The Artificial Self-Replication Page,
maintained by Moshe Sipper, summarizes research on self-replicating
systems arranged in chronological order. Each system is described
by title, author(s), year, model, implementation, and a short
description.
- The
Complexity and Artificial Life Research Web site,
maintained by the non-profit organization CalResCo, is dedicated to the
promotion of the Complexity System Sciences.
- The
Alife Models of Flocking Behavior
site maintained by Craig Reynolds, famous for his "boids," contains
extensive references and pointers to work related to group motion in
animals.
- The
Web Page on Individual-Based Models
is an annotated list of links maintained by Craig Reynolds with a
stress on individual-based modelling using spatially explicit mobile
agents in continuous space.
-
The Institute for Information Technology, under the auspices of
Canada's National Research Council, maintains a Web site of
Artificial Intelligence Resources.
- The goal of
BotSpot
is to be a definitive resource for bots, intelligent agents, and
artificial intelligence on the Web. The site includes: fourteen
searchable Bot Classification Databases for bot implementations;
FAQs; book recommendations; and pointers to articles, electronic
journals, upcoming conferences, previous conference proceedings, and
language and code for creating bots and intelligent agents. The
site also supports a free monthly BotSpot Newsletter. BotSpot has
received over 150 awards in its first twelve months of operation,
including a designation by PC Magazine as one of their top
100 recommended Web sites.
Reminder: News Items Requested for ACE News Notes and Complexity
Just a reminder to send me (tesfatsi@iastate.edu) any news items that
you would like to have considered for inclusion in the ACE news notes and/or
the Complexity-at-Large section of the John Wiley journal Complexity.
Thanks.
Copyright © 1998 Leigh Tesfatsion. All Rights Reserved.