## Anton Klimovsky

I am a lecturer at the University of Duisburg-Essen and a part of the Probability Theory group there.

In the Summer Semester 2014, I was an interim professor for Probability and Statistics at the University of Bonn.

Previously, I was a postdoctoral fellow at the Leiden University (group of Prof. Dr. F. den Hollander), Eindhoven University of Technology (EURANDOM, supported by a Mary Curie Postdoctoral Fellowship) and at the University of Erlangen-Nürnberg (group of Prof. Dr. A. Greven).

I received my Ph.D. (Dr.rer.nat.) from Berlin University of Technology in 2008 and a Diploma in Applied Mathematics from Kharkiv National University in 2002.

## Mathematical interests

My research interests are concentrated around stochastic models of complex systems (i.e., systems of many interacting elements). In particular, I work on statistical mechanics of disordered systems (i.e., study of global/collective behaviour of complex systems with highly inhomogeneous interactions) and especially on range-free spin glasses. In this realm, a recurring theme of my research concerns the asymptotic behaviour of large sums of strongly correlated exponentials, as both the number of their summands and the effective dimension of the correlation structure simultaneously tend to infinity. The goal here is to identify the new universal limiting probabilistic structures (and their domains of attraction) that appear in systems with strongly interacting parts.

In parallel, I am interested in the models of evolution for spatially extended populations and, in particular, on stochastic models for population genetics. One prominent goal here is to identify universal spatio-temporal patterns (especially the hierarchical ones) that emerge in such models.

High-level heuristic ideas originating in statistical physics (such as the cavity method and the renormalization group methodology) in combination with rigorous probabilistic techniques such as

- weak convergence theory;
- extreme value theory;
- large deviations;
- probabilistic inequalities and concentration of measure;
- probabilistic methods for non-linear PDEs;

play important role in tackling the above mentioned problems. (Though, an exciting and very challenging aspect here is that many of the standard probabilistic techniques break down and one has to develop the new ones!)

### Contact information

Sprechstunde (während der Vorlesungszeit): Montags, 14:00-15:00 Uhr

anton.klymovskiy@uni-due.de

phone: +49 201 183 6714

fax: +49 201 183 2381

**Mailing address:**

Universität Duisburg-Essen

Fakultät für Mathematik

AG Wahrscheinlichkeitstheorie

Dr. Anton Klymovskiy

D-45117 Essen

Germany

**Visiting address:**

Universität Duisburg-Essen

Fakultät für Mathematik

Room WSC-W-3.15

Thea-Leymann-Str. 9

D-45127 Essen