The aim of this dissertation is to investigate ergodic properties, in particular unique ergodicity, in a noncommutative setting, that is in C*-dynamical systems. Fairly recently Abadie and Dykema introduced a broader notion of unique ergodicity, namely relative unique ergodicity. Our main focus shall be to present their result for arbitrary abelian groups containing a F lner sequence, and thus generalizing the Z-action dealt with by Abadie and Dykema, and also to present examples of C*-dynamical systems that exhibit variations of these (uniquely) ergodic notions. Abadie and Dykema gives some characterizations of relative unique ergodicity, and among them they state that a C*-dynamical system that is relatively uniquely ergodic has a conditional expectation onto the xed point space under the automorphism in question, which is given by the limit of some ergodic averages. This is possible due to a result by Tomiyama which states that any norm one projection of a C*-algebra onto a C*-subalgebra is a conditional expectation. Hence the rst chapter is devoted to the proof of Tomiyama's result, after which some examples of C*-dynamical systems are considered. In the last chapter we deal with unique and relative unique ergodicity in C*-dynamical systems, and look at examples that illustrate these notions. Speci cally, we present two examples of C*-dynamical systems that are uniquely ergodic, one with an R2-action and the other with a Z-action, an example of a C*-dynamical system that is relatively uniquely ergodic but not uniquely ergodic, and lastly an example of a C*-dynamical system that is ergodic, but not uniquely ergodic. / Dissertation (MSc)--University of Pretoria, 2013. / gm2014 / Mathematics and Applied Mathematics / unrestricted

Cette thèse est constituée de deux parties indépendantes. La première partie concerne l'étude des vecteurs propres de matrices aléatoires de type Wigner. Dans un premier temps, nous étudions la distribution des vecteurs propres de matrices de Wigner déformées, elles consistent en une perturbation d'une matrice de Wigner par une matrice diagonale déterministe. Si les deux matrices sont du même ordre de grandeur, il a été prouvé que les vecteurs propres se délocalisent complètement et les valeurs propres rentrent dans la classe d'universalité de Wigner-Dyson-Mehta. Nous étudions ici une phase intermédiaire où la perturbation déterministe domine l'aléa: les vecteurs propres ne sont pas totalement délocalisés alors que les valeurs propres restent universelles. Les entrées des vecteurs propres sont asymptotiquement gaussiennes avec une variance qui les localise dans une partie explicite du spectre. De plus, leur masse est concentrée autour de cette variance dans le sens d'une unique ergodicité quantique. Ensuite, nous étudions des corrélations de différents vecteur propres. Pour se faire, une nouvelle observable sur les moments de vecteurs propres du mouvement brownien de Dyson est étudiée. Elle suit une équation parabolique close qui est un pendant fermionique du flot des moments de vecteurs propres de Bourgade-Yau. En combinant l'étude de ces deux observables, il est possible d'analyser certaines corrélations.La deuxième partie concerne l'étude de la distribution des valeurs propres de modèles non-linéaires de matrices aléatoires. Ces modèles apparaissent dans l'étude de réseaux de neurones aléatoires et correspondent à une version non-linéaire de matrice de covariance dans le sens où une fonction non-linéaire, appelée fonction d'activation, est appliquée entrée par entrée sur la matrice. La distribution des valeurs propres convergent vers une distribution déterministe caractérisée par une équation auto-consistante de degré 4 sur sa transformée de Stieltjes. La distribution ne dépend de la fonction que sur deux paramètres explicites et pour certains choix de paramètres nous retrouvons la distribution de Marchenko-Pastur qui reste stable après passage sous plusieurs couches du réseau de neurones. / This thesis consists in two independent parts. The first part pertains to the study of eigenvectors of random matrices of Wigner-type. Firstly, we analyze the distribution of eigenvectors of deformed Wigner matrices which consist in a perturbation of a Wigner matrix by a deterministic diagonal matrix. If the two matrices are of the same order of magnitude, it was proved that eigenvectors are completely delocalized and eigenvalues belongs to the Wigner-Dyson-Mehta universality class. We study here an intermediary phase where the deterministic perturbation dominates the randomness of the Wigner matrix : eigenvectors are not completely delocalized but eigenvalues are still universal. The eigenvector entries are asymptotically Gaussian with a variance which localize them onto an explicit part of the spectrum. Moreover, their mass is concentrated around their variance in a sense of a quantum unique ergodicity property. Then, we consider correlations of different eigenvectors. To do so, we exhibit a new observable on eigenvector moments of the Dyson Brownian motion. It follows a closed parabolic equation which is a fermionic counterpart of the Bourgade-Yau eigenvector moment flow. By combining the study of these two observables, it becomes possible to study some eigenvector correlations.The second part concerns the study of eigenvalue distribution of nonlinear models of random matrices. These models appear in the study of random neural networks and correspond to a nonlinear version of sample covariance matrices in the sense that a nonlinear function, called the activation function, is applied entrywise to the matrix. The empirical eigenvalue distribution converges to a deterministic distribution characterized by a self-consistent equation of degree 4 followed by its Stieltjes transform. The distribution depends on the function only through two explicit parameters. For a specific choice of these parameters, we recover the Marchenko-Pastur distribution which stays stable after going through several layers of the network.

No description available.

We consider the incompressible 2D Navier-Stokes equations on the torus driven by a deterministic time quasi-periodic force and a noise that is white in time and extremely degenerate in Fourier space. We show that the asymptotic statistical behavior is characterized by a uniquely ergodic and exponentially mixing quasi-periodic invariant measure. The result is true for any value of the viscosity ν > 0. By utilizing this quasi-periodic invariant measure, we show the strong law of large numbers and central limit theorem for the continuous time inhomogeneous solution processes. Estimates of the corresponding rate of convergence are also obtained, which is the same as in the time homogeneous case for the strong law of large numbers, while the convergence rate in the central limit theorem depends on the Diophantine approximation property on the quasi-periodic frequency and the mixing rate of the quasi-periodic invariant measure. We also prove the existence of a stable quasi-periodic solution in the laminar case (when the viscosity is large). The scheme of analyzing the statistical behavior of the time inhomogeneous solution process by the quasi-periodic invariant measure could be extended to other inhomogeneous Markov processes.