This site aims to be a comprehensive repository of informations on the
Iterated Prisoner's Dilemma, and furthermore on the representation, study
and knowledge of cooperation (and evolution of cooperation) between
agents.
Stuff presented here has been collected by the SMAC team,
which belong to the LIFL, as
part of a project called PRISON. The goal of the project
being the use of computer science tools in order to collect informations on
the way cooperation between autonomous agents could arise and be maintaned.
On this site you will find a presentation of the Classical Iterated Prisoner's
Dilemma (CIPD), SMAC published papers as well as a bibliography about the
prisoner's dilemma, simulation softwares, available for downloading as well as
for online testing, and some links of close interest.
Keywords: (Iterated) Prisoner's Dilemma,
Cooperation, Artificial Life, Evolution, Game Theory, Multi-Agents
Systems
- [2004/11/02]
- We completely disagree with the way prisoner's dilemma competition at
CEC'2004 has been done as well as the way it will be done at CIG'05. You can
read and sign
our statement on this competition and the dissemination of
results some participant try to attribute themself from the results
of this competition.
- [1999/11/09]
- A prisoner's dilemma competition is organized at CEC'2000 . We
encourage you to participate !
- [1998/11/26]
- Added the sources of the classes of jprison in the distributions..
- [1998/11/20]
- Added the jprison package.
- Added the first three applets using jprison.
- [1998/11/19]
- Added the description of basic strategies.
- [1998/09/25]
- The site is open on the new adress.
- [1998/07/01]
- The preliminary applet
versions are available.
Let two artificial rational agents have the choice between two moves:
- Cooperation, let us note C
- Defection, let us note D
They play one against the other, in a synchronous manner, so that they do
not know what the other will play. They get a score according to the situation
of the move:
- They both cooperate and then get both the Reward for
cooperation payoff, let evaluate it to R points;
- They both defect and then get both the selfish Punishment
payoff, let it be P points;
- One chooses to defect while the other chooses to cooperate, then the one
who has defected gets the Temptation payoff, let it be
T points, and the one who has cooperated gets the
Sucker's score, let it be S points.
The classical choice of value for payoff is (row player payoff are given
first):
| Cooperate | Defect |
| Cooperate |
R = 3
R = 3 |
S = 0
T = 5 |
| Defect |
T = 5
S = 0 |
P = 1
P = 1 |
To have a dilemma, temptation must pay more than cooperation, which must
pay more than punishment, which must be better than to be the sucker. This is
formalised by:
T > R > P > S
Since the one-shot version of the Prisoner's Dilemma is not very
interesting (the most rational choice is to defect), the game is repeated an
unknown number of times.
The game is said iterated.
The final score of a payer is the sum of all its moves score.
Since none player knows when the game will be ended, it is possible to study
each agent's strategy, to look, for instance, how each player tries to put
cooperation in the game.
In order to favour cooperation, i.e. common interest at the expens of
selfish interest, this inequation is respected:
2 R > T + S
With this restriction, strategies have no advantage in alternatively
cooperate and defect. To study the behavior of strategies, two kinds of
computation can be done.
- The first one is a simple round-robin tournament, in
which each strategy meets all other strategies. Its final score is then
the sum of all scores done in each confrontation. At the end, the
strategy's strength measurement is given by its rank in the tournament.
- The second one is a simulated ecological evolution, in
which at the beginning there is a fixed population including the same
quantity of each strategy. A round-robin tournament is made and then the
population of bad strategies is decreased whereas good strategies obtain
new elements. The simulation is repeated until the population has been
stabilised (the population does not change anymore). This is this way
that nasty strategies, those who take the initiative of the first
defection, have been discovered to be not very stable, because they are
invaded by kind ones.
Here is a description of some of the basic strategies used in our simulations
as well as in the literature:
- all_c
- Always cooperates. [c]*
- all_d
- Always defects. [d]*
- tit_for_tat
- The tit_for_tat strategy was introduced by Anatole Rapoport. It begins
to cooperate, and then play what its opponent played in the last move.
- spiteful
- It cooperates until the opponent has defected, after that move it always
defects.
- soft_majo
- Plays opponent's majority move, if equal then cooperates. First move is
considered to be equality.
- per_ddc
- Plays periodically : [d,d,c]*
- per_ccd
- Plays periodically : [c,c,d]*
- mistrust
- Defects, then plays opponent's move.
- per_cd
- Plays periodically [c,d].
- pavlov
- The win-stay/lose-shift strategy was introduced by Martin Nowak and Karl
Sigmund. It cooperates if and only if both players opted for the same
choice in the previous move.
- tf2t
- Cooperates except if opponent has defected two consecutive times.
- hard_tft
- Cooperates except if opponent has defected at least one time in the two
previous move.
- slow_tft
- Plays [c,c], then if opponent plays two consecutive time the same move
plays its move.
- hard_majo
- Plays opponent's majority move, if equal then defects. First move is
considered to be equality.
- random
- Cooperates with probability 1/2.
As for today our team is composed by people working in the SMAC team of
the LIFL:
Other peoples are, or has been, involved in the development of our
works:
- Y. Leborgne, undergraduate student, is involved in some
mathematicals research on a variation of the Iterated Prisoner's
Dilemma, we have set up.
- F. Laveine, graduate student, has been involved in the
construction of the Xview interface to the PRISON software;
- F. Grinion, graduate student, has been involved in the
construction of the WINPRI software;
Here are an extensive list of the papers written, and published by members of
the team. They mainly present our ideas and results about the cooperation
through the IPD model.
- L'Altruisme perfectionné
Jean-Paul DELAHAYE, Philippe MATHIEU
Publication interne du LIFL, numéro IT-249
- Expériences sur le Dilemme Itéré des
Prisonniers [ps.gz]
Jean-Paul DELAHAYE, Philippe MATHIEU
Publication Interne du LIFL, numéro IT-233
An automatically generated complete bibliography of all papers or books we
have is available here.
An annotated bibliography on the Evolution of Cooperation has been set up by
Robert Axelrod and Lisa
D'Ambrosio, and is available here.
The original file is at http://pscs.physics.lsa.umich.edu/RESEARCH/Evol_of_Coop_Bibliography.html.
You can try to learn the model a little bit more deeply by trying some of our
online applets.
In order to may be able to run it you have to ensure your browser is
able to run JAVA program constructed with the JDK 1.1.*, and is able to render
AWT objects
For instance Netscape browser are able of that only since release
4.06, so that our applets will not work only with previous
releases.
At that time you may try three applets :
- The first one enables you to try simple meeting of two strategies
playing to the Classical Iterated Prisoner's Dilemma, in order to
understand the way strategies acts. Click here to try it.
- The second one enables you to test strategies, playing to the Classical
Iterated Prisoner's Dilemma, with tournaments, or championship, so that
you can appreciate the basic way of evaluating a particuliar
strategy. Click here to try it.
- The third one enables you to test the evolution of population of agents
playing any two by two game you may imagine and define through the payoff
matrix. You may choose the size of each population, view tournament
results as well as a graphical results of the population
evolution. Click here to try it.
We have produced a lot of simulation software, since the beginning of our
studies of the dilemma.
Here is the description of each of those software projects:
- WINPRI
-
- This project aims to define a user friendly way of working
on the Iterated Prisoner's Dilemma. The purpose is to
allow all kind of people to make simulations on
cooperation and evolution of cooperation. Strategies can
be choosen in a predefined set, set-up by customizing
generic frame. This tool is perfect for non
specialist, and thus for lecture on the IPD or other two
player games (Game Theory meaning) project.
- You can exclusively use it on Windows (3.x, 95 or NT). The
program is written in Visual Basic 3.0 and is thus not as
fast as the PRISON project.
- All operations are made graphically with the mouse (screenshot).
- Download
winpri.zip (315kb):
- PRISON
-
- This project aims to define an optimized, powerful and
efficient way, to make big experimentations, as we do, on
the Iterated Prisoner's Dilemma. It is written entirely in
ANSI C. It allows you to write any possible strategy using
the expression power of the C language. Use it only if
you make big experimentations or if you need speed or
special strategies.
- You can use it on any platform with an ANSI-C compiler,
needed to modify, or create new strategies. If you want to
see graphics interpretations of your results you will need
to have GNUPLOT installed on your system, with which it is
interfaced, otherwise you will have to treat yourself the
results which are stored in data text file.
- All operations are made in text and command line mode, but
on UNIX platforms, using X/Window and the Xview toolkit,
you can use a graphical interface called xprison (screenshot). Read the README file before
downloading any archives.
- Download it:
prison.tar.gz (370kb)
prison.zip (390kb)
- JPRISON
-
- The goal of this project is to offer JAVA
classes which could be used to easily create Iterated
Prisoner's Dilemma simulations. It is clearly devoted to
people with some JAVA programming skills, but are simple
enough for beginners to use. All kind of strategies may be
user implemented. The simulations are open to all two by
two games, i.e. with a different payoff matrix than the
one of the IPD. This tool is perfect for programming
lecture, as well as for people wanted to design their own
simulations. It is a library not a
program.
- You can use it on any platform with a Java Virtual Machine
available. The classes are entirely written in JAVA, and
have been tested with the Sun JDK
1.1.x, with x > 5. The power of the
simulations created are tributed to the power of the Java
Virtual Machine used.
- Some examples applets are distributed with the archive and
are available on this site in the previous section. Read the README file
before downloading any archives. Full sources and API
documentation are also distributed in the
archives. Documentation is also available online here.
- Download it:
jprison.tar.gz (116kb)
jprison.zip (202kb)
EVOLUTION
- This project aims to help a user to compute ecological
evolution based on a defined payoff matrix involving only
3 strategies. The computation may be done in 3 differents
ways (using integers, float, etc.)
- This is a Micro$oft Excel 2000 sheet using some macros.
- All operations are made via the Mico$oft Excel 2000
software tool.
- Download it:
|
It is although to be noticed that a new product, based on the ideas of
WinPri, and on the classes of jprison is at work. It will be available
as soon as the site will be updated.. So check the page often to have
more informations.
You can join us by e-mail at : prison@lifl.fr.
Otherwise we could be joined by snail mail at this address :
Philippe Mathieu - PRISON project headmaster
Laboratoire d'Informatique Fondamentale de Lille
Université des Sciences et Technologies de Lille
UFR IEEA - Bât M3
59655 Villeneuve d'Ascq Cedex
FRANCE
|
We could also be joined by phone or fax at those numbers :
|
Phone
|
Fax
|
|
P. Mathieu :
|
+33 3 20 43 45 04
|
|
B. Beaufils :
|
+33 3 20 43 69 02
|
|
+33 3 20 43 65 66
|
Some links on the subject are collected here.
The PRISON project is sponsored by :
This page has been accessed 29713 times since
Friday, 25-Sep-98 17:32:21 MET DST
