Victor Poupet

Directional dynamics along arbitrary curves in cellular automata

This paper studies directional dynamics on one-dimensional cellular automata, a formalism previously introduced by the third author. The central idea is to study the dynamical behavior of a cellular automaton through the conjoint action of its global rule (temporal action) and the shift map (spacial action): qualitative behaviors inherited from topological dynamics (equicontinuity, sensitivity, expansivity) are thus considered along arbitrary curves in space-time. The main contributions of the paper concern equicontinuous dynamics which can be connected to the notion of consequences of a word. We show that there is a cellular automaton with an equicontinuous dynamics along a parabola, but which is sensitive along any linear direction. We also show that real numbers that occur as the slope of a limit linear direction with equicontinuous dynamics in some cellular automaton are exactly the computably enumerable numbers.

On the complexity of limit sets of cellular automata associated with probability measures

We study the notion of limit sets of cellular automata associated with probability measures (μ-limit sets). This notion was introduced by P. Kůrka and A. Maass in [1]. It is a refinement of the classical notion of ω-limit sets dealing with the typical long term behavior of cellular automata. It focuses on the words whose probability of appearance does not tend to 0 as time tends to infinity (the persistent words). In this paper, we give a characterization of the persistent language for non sensitive cellular automata associated with Bernoulli measures. We also study the computational complexity of these languages. We show that the persistent language can be non-recursive. But our main result is that the set of quasi-nilpotent cellular automata (those with a single configuration in their μ-limit set) is neither recursively enumerable nor co-recursively enumerable.

Cellular automata: real-time equivalence between one-dimensional neighborhoods

It is well known that one-dimensional cellular automata working on the usual neighborhood are Turing complete, and many acceleration theorems are known. However very little is known about the other neighborhoods. In this article, we prove that every one-dimensional neighborhood that is sufficient to recognize every Turing language is equivalent (in terms of real-time recognition) either to the usual neighborhood {–1,0,1} or to the one-way neighborhood {0,1}.

Real time language recognition on 2D cellular automata: dealing with non-convex neighborhoods

In this paper we study language recognition by twodimensional cellular automata on different possible neighborhoods. Since it is known that all complete neighborhoods are linearly equivalent we focus on a natural sub-linear complexity class: the real time. We show that any complete neighborhood is sufficient to recognize in real time any language that can be recognized in real-time by a cellular automaton working on the convex hull of V.

Asymptotic Cellular Complexity

We show here how to construct a cellular automaton whose asymptotic set (the set of configurations it converges to) is maximally complex: it contains only configurations of maximal Kolmogorov complexity. This cellular automaton hence exhibits the most complex possible asymptotic behavior.

A padding technique on cellular automata to transfer inclusions of complexity classes

We will show how padding techniques can be applied on onedimensional cellular automata by proving a transfer theorem on complexity classes (how one inclusion of classes implies others). Then we will discuss the consequences of this result, in particular when considering that all languages recognized in linear space can be recognized in linear time (whether or not this is true is still an open question), and see the implications on one-tape Turing machines.

Simulating 3D Cellular Automata with 2D Cellular Automata

The purpose of this article is to describe a way to simulate any 3-dimensional cellular automaton with a 2-dimensional cellular automaton. We will present the problems that arise when changing the dimension, propose solutions and discuss the main properties of the obtained simulation.

Cellular Automata: Real-Time Equivalence between One-Dimensional Neighborhoods

It is well known that one-dimensional cellular automata working on the usual neighborhood are Turing complete, and many acceleration theorems are known. However, very little is known about the other neighborhoods. In this article we prove that every one-dimensional neighborhood that is big enough so that every letter of the entry word can affect the computation is equivalent (in terms of real-time recognition) either to the usual neighborhood {-1,0,1} or to the one-way neighborhood {0,1}.

A Linear Acceleration Theorem for 2D Cellular Automata on all Complete Neighborhoods

Linear acceleration theorems are known for most computational models. Although such results have been proved for two-dimensional cellular automata working on specific neighborhoods, no general construction was known. We present here a technique of linear acceleration for all two-dimensional languages recognized by cellular automata working on complete neighborhoods.

Comparing 1D and 2D Real Time on Cellular Automata

We study the influence of the dimension of cellular automata (CA) for real time language recognition of one-dimensional languages with parallel input. Specifically, we focus on the question of determining whether every language that can be recognized in real time on a 2-dimensional CA working on the Moore neighborhood can also be recognized in real time by a 1-dimensional CA working on the standard two-way neighborhood. We show that 2-dimensional CA in real time can perform a linear number of simulations of a 1-dimensional real time CA. If the two classes are equal then the number of simulated instances can be polynomial.