ANR project Aléatoire, Dynamique et
Spectre (ADYCT)
Aims of the research project.
The last fifteen years have witnessed several significant progress in the
analytical understanding of chaotic dynamical systems, and at the
same time in the study of models of random waves. Even if they
took place in parallel, these new developments share many similarities in
their objectives and methods: asymptotic distribution of eigenvalues,
existence of spectral gaps, decay of correlations, etc. Both topics play a
fundamental role in the study of quantized chaotic sytems, a domain
colloquially referred to as quantum chaos. The aim of this
proposal is to pursue the development of these topics, and develop the
growing interactions between them. This will be achieved by focusing on
three related tasks:
- Random waves, hyperbolic dynamics and
quantum chaos. In this first task, we will deepen our
understanding of the random wave model, in particular with regard to
applications to quantum chaos. We will also analyze the ergodic
properties of classical chaotic systems, through the scope of their
spectrum of Ruelle resonances, mimicking the analysis of quantum
systems.
- Weakly chaotic systems, cohomological
equations and differential topology. In this second task, we will
extend the study of Ruelle spectrum to models with weaker chaotic
properties, such as Teichmüller flows or Axiom A flows. We will also
elaborate on the recent developments about the topological content of
this Ruelle spectrum.
- Spectra of random operators.In this
last task, we will group operators within probabilistic ensembles,
thereby defining classes of random operators, and study their spectral
properties from a probabilistic viewpoint. In particular, we will show
how this randomness can help to address some questions raised in the
first two tasks, e.g. existence of spectral gaps, asymptotic
distribution of eigenvalues, properties of eigenfunctions, etc.