In the first part of this thesis we transfer a result of Guillopé et al. concerning the number of zeros of the Selberg zeta function for convex cocompact Schottky groups to the setting of certain types of graph directed Markov systems (GDMS). For these systems the zeta function will be a type of Ruelle zeta function. We show that for a finitely generated primitive conformal GDMS S, which satisfies the strong separation condition (SSC) and the nestedness condition (NC), we have for each c>0 that the following holds, for each w \in\$C$ with Re(w)>-c, |\Im(w)|>1 and for all k \in\$N$ sufficiently large: log | zeta(w) | <<e {delta(S).log(Im|w|)} and card{w \in\ Q(k) | zeta(w)=0} << k {delta(S)}. Here, Q(k)\subset\%C$ denotes a certain box of height k, and delta(S) refers to the Hausdorff dimension of the limit set of S. In the second part of this thesis we show that in any dimension m \in\$N$ there are GDMSs for which the Hausdorff dimension of the uniformly radial limit set is equal to a given arbitrary number d \in\(0,m) and the Hausdorff dimension of the Jørgensen limit set is equal to a given arbitrary number j \in\ [0,m). Furthermore, we derive various relations between the exponents of convergence and the Hausdorff dimensions of certain different types of limit sets for iterated function systems (IFS), GDMSs, pseudo GDMSs and normal subsystems of finitely generated GDMSs. Finally, we apply our results to Kleinian groups and generalise a result of Patterson by showing that in any dimension m \in\$N$ there are Kleinian groups for which the Hausdorff dimension of their uniformly radial limit set is less than a given arbitrary number d \in\ (0,m) and the Hausdorff dimension of their Jørgensen limit set is equal to a given arbitrary number j \in\ [0,m).

Les automates cellulaires sont à la fois un modèle de calcul parallèle, un système complexe et un système dynamique. Ils fonctionnent de manière synchrone et en temps discret, leur particularité est que les fonctions qu'ils définissent sont issues de l'application simultanée, en tout point de l'espace, d'une règle d'évolution locale. L'ensemble limite est un objet classique des systèmes dynamiques, c'est l'ensemble des états que le système peut atteindre arbitrairement tard. Il a été très étudié dans le cadre des automates cellulaires, et les résultats sont nombreux. Parmi ces résultats, un théorème de Rice démontré par Jarkko Kari dit que toute propriété des ensembles limites est indécidable. Dans ce mémoire, on ne s'intéresse plus à l'ensemble limite traditionnel, mais à une variante pour laquelle on utilise une mesure sur l'espace des entrées, sélectionnant ainsi les comportements susceptibles d'apparaître arbitrairement tard et souvent. Ce nouvel ensemble, que l'on nomme ensemble mu-limite, a été introduit en 2000 par Petr Kurka et Alejandro Maass. La plupart des résultats sur les ensembles limites ne se transposent pas naturellement. On étudie la famille des ensembles mu-limites d'automates cellulaires. On montre que sous certaines contraintes sur la dynamique, l'ensemble mu-limite peut être entièrement décrit. On classe ainsi les automates en fonction de ces contraintes. Dans le cas général, on montre l'existence d'automates cellulaires ayant comme ensembles mu-limites un grand nombre d'ensembles complexes. On finit par montrer un théorème de Rice pour les ensembles mu-limites d'automates cellulaires: tout propriété non triviale de ces ensembles est indécidable. / Cellular automata are simultaneously a model of parallel computation, a complex system and a dynamical system. They are synchronous and time is discrete. The functions defined by their application is the result of the synchronous application of the same local rule everywhere. The limit set is a classical tool of dynamical systems theory, it is the set of states the system can reach arbitrarily late. It has been studied often in the particular case of cellular automata and there are numerous results. Amongst them, a Rice's theorem proved by Jarkko Kari states that any non-trivial property of limit sets of cellular automata is undecidable. In this thesis, we do not consider the classical limit set, as we add a measure on the space of states of the system. Thus, we get a set which contains behaviors that appear arbitrarily far and often. This set is named mu-limit set and was introduced in 2000 by Petr Kurka and Alejandro Maass. Most of the results on limit sets cannot be directly adapted for mu-limit sets. We study the family of all mu-limit sets of cellular automata. We show that under some constraints on the dynamics, the mu-limit set can be entirely described. We then produce a classification of cellular automata according to these constraints. In the general case, we prove the existence of cellular automata whose mu-limit sets are among a large set of complex sets. We finally prove Rice's theorem for mu-limit sets: any non-trivial property is undecidable.

In this dissertation we study graph directed Markov systems (GDMS) and limit sets associated with these systems. Given a GDMS S, by the Hausdorff dimension spectrum of S we mean the set of all positive real numbers which are the Hausdorff dimension of the limit set generated by a subsystem of S. We say that S has full Hausdorff dimension spectrum (full HD spectrum), if the dimension spectrum is the interval [0, h], where h is the Hausdorff dimension of the limit set of S. We give necessary conditions for a finitely primitive conformal GDMS to have full HD spectrum. A GDMS is said to be regular if the Hausdorff dimension of its limit set is also the zero of the topological pressure function. We show that every number in the Hausdorff dimension spectrum is the Hausdorff dimension of a regular subsystem. In the particular case of a conformal iterated function system we show that the Hausdorff dimension spectrum is compact. We introduce several new systems: the nearest integer GDMS, the Gauss-like continued fraction system, and the Renyi-like continued fraction system. We prove that these systems have full HD spectrum. A special attention is given to the backward continued fraction system that we introduce and we prove that it has full HD spectrum. This system turns out to be a parabolic iterated function system and this makes the analysis more involved. Several examples have been constructed in the past of systems not having full HD spectrum. We give an example of such a system whose limit set has positive Lebesgue measure.

Nous étudions les semi-groupes de matrices avec des points de vue variés qui se re-coupent. Le point de vue de la croissance s’avère relié à un point de vue géométrique : nous avons partiellement généralisé aux semi-groupes un théorème de Patterson-Sullivan-Paulin sur les groupes, qui donne l’égalité entre exposant critique et dimension de Hausdorff de l’ensemble limite. Nous obtenons cela dans le cadre général des semi-groupes d’isométries d’un espace Gromov-hyperbolique, et notre preuve nous a permis d’obtenir également d’autres résultats nouveaux. Le point de vue informatique s’avère également relié à la croissance, puisque la notion de semi-groupe fortement automatique, que nous avons introduit, permet de calculer les exposants critiques exactes de semi-groupes de développement en base β. Et ce point de vue donne également beaucoup d’autres informations sur ces semi-groupes. Cette notion de croissance s’avère aussi reliée à des conjectures sur les fractions continues telles que celle de Zaremba. Et c’est en étudiant certains semi-groupes de matrices que nous avons pu démontrer des résultats sur les fractions continues périodiques bornées qui permettent de petites avancées dans la résolution d'une conjecture de McMullen. / We study matrix semigroups with different point of view that overlaps. The growth point of view seems to be related with the geometric point of view : we partially generalize to the semigroups a theorem on groups of Patterson-Sullivan-Paulin, that give the equality between the critical exponent and the Hausdorff dimension of the limit set. We obtain this in the general framework of isometries of a Gromov-hyperbolic space, and our proof give also others new results. The computer science point of view is also related to the growth, since we obtain a way to calculate exact values of critical exponents of somes β-adic development semigroups, from a notion of automatic semigroups that we introduce. Furthermore this point of view give a lot of information on these semigroups. This notion of growth shows to be also related to conjectures on continued fractions like Zaremba’s one. And by studing some matrix semigroups we were able to prove some results on bounded periodic continued fractions, doing a little step in the resolution of a conjecture of McMullen.