home research various publications
Zao Wou Ki - Coll. E.N.S Lyon - 1987
Zao Wou Ki - Sans Titre - 1987

« Je m'en tiens à mes limites; elles sont plus fécondes que ce qui les excède. »
Philippe Jaccottet, Nuages (éd. Fata Morgana)

Cellular automata

Some recent work

  • A collaboration with Biswanatah Sethi and Sukanata Das (Univ. of West Bengal, India) has appeared in a book on reversible computing: see On the reversibility of ECAs with fully asynchronous updating: the recurrence point of view [preprint]. This publication is edited by A. Adamatzky for K. Morita's 80th birthday.
  • I am also interested in the study of the global synchronisation problem in cellular automata. The problem is to make a system converge to a homogeneous state where all the cells blink simultaneously. A communication on this topic was presented at the AUTOMATA'15 conference and a long version appeared in Natural computing.

Density classification problem

I have worked Irène Marcovici, Ana Bušić and Jean Mairesse on the study of the density classification problem (see this paper) on infinite lattices and graphs.

With Irène Marcovici and Siamak Taati, we proposed an interacting-particle system that solves the density classification problem in the case of finite two-dimensional grids (tori). A communication was presented at the AUTOMATA'16 conference

Phase transitions in Cellular Automata

Changing the probability of a transition in asynchronous cellular automata triggers phase transitions. -> See: Asynchronism Induces Second Order Phase Transitions in Elementary Cellular Automata, Journal of Cellular Automata, Vol. 4, No. 1, 2009.
Click here to access to additional numerical results and figures.

I worked with Hugues Berry on an experimental determination of phase transitions in discrete reaction-diffusion systems.
-> See: Robustness of the Critical Behaviour in a Discrete Stochastic Reaction-diffusion Medium
Click here to access additional data.

Asynchronous updating of 2D CA

With Joerg Hoffmann, we examined planning techniques to (a) plan a sequence of update in order to reach a fixed point in some 2D binary CAs (b) automatically proove that such points are reachable.
-> See: "Brothers in Arms? On AI Planning and Cellular Automata", in Proceedings of the 19th European Conference on Artificial Intelligence (ECAI'10), Lisbon, Portugal, August 2010.

I had collaboration with Lucas Gerin on the study of asynchronous cellular automata in two dimensions. We examined the convergence properties to a fixed poind and studied some simple rules as a first step for understanding these CA.
-> See: Examples of Fast and Slow Convergence of 2D Asynchronous Cellular Systems, Journal of Cellular Automata, Vol. 4, No. 4, 2009.

Contacts in Canada (summer 2007 & 2010)

We started exchanging with Henryk Fuks to examine both experimentally and analytically which are the asynchronous CA that conserve various quantities. In a frist step these quantities are the density, the ratio of 01 or 11 patterns, etc.

Experimental results for the 88 ECA: Synthesis Density Kinks Block11

I was also hosted by Paola Flocchini. The Fuzzy CA model she studies was included into the FiatLux software.

Technè - Art - Computation

With the contemporary artists René Sultra and Maria Barthélémy, we generated rugs with cellular automata patterns. The carpets were exposed in a art exhibition held at "Maison Salvan", in Carjac, near Toulouse, South of France, in June 2008. They were created with the FiatLux CA simulator. We used Elementary Cellular Automata with random initial conditions. Photos (click to enlarge):
expo Barthelmy Sultra image 7486 expo Barthelmy Sultra image 7516
expo Barthelmy Sultra image 7495 expo Barthelmy Sultra image 7533

Carton d'invitation de l'exposition

Multi-agent systems

Assiette au musée du Louvre (c) Site du musée
inscription : La science, son goût est amer au début, mais à la fin plus doux que le miel. La santé [au possesseur].
département arts de l'Islam du musée du Louvre.

The multi-turmite system

This joint work with Vincent Chevrier. We are revisiting the Influence/Reaction formalism proposed by Ferber and Muller. Our main objetcive is to examine how to describe a simple multi-agent system, namely the muti-turmite model (or Langton's ants), as a discret dynamical system. Click here to learn more.

Selma Belgacem wrote her Master's thesis on the study of an asynchronous version of the multi-Turmite system, also known as the multiple Langton ants system. See here for a paper linked to this research.

Decentralised gathering with local rules

We are studying how to group autonomous agents that move on discrete field and have only a short view of their environment.

From Multi-agent Systems to Cellular Automata

This was a joint work with Olivier Simonin and Antoine Spicher. We examined how to "translate" the individual-based language of multi-agent systems into the cell-centered language of cellular automata. See this paper

Last Update : Feb. 2020