Dean — Ivan Arzhantsev
First Deputy Dean — Tamara Voznesenskaya
Deputy Dean for Research and International Relations — Sergei Obiedkov
Deputy Dean for Methodical and Educational Work — Ilya Samonenko
Deputy Dean for Development, Finance and Administration — Irina Plisetskaya
Phone: +7 (495) 772-95-90 * 12332
125319, Moscow, 3 Kochnovsky Proezd (near metro station 'Aeroport').
Bienvenu M., Kikot S., Kontchakov R. et al.
Journal of the ACM. 2018. Vol. 65. No. 5. P. 28:1-28:51.
Doikov Nikita, Richtarik P.
Proceedings of Machine Learning Research. 2018. No. 80. P. 1290-1298.
Hushchyn M., Chekalina V.
Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment. 2018. P. 1-2.
Naumov A., Spokoiny V., Ulyanov V. V.
Probability Theory and Related Fields. 2019. P. 1-42.
Shapoval A., Le Mouël J., Shnirman M. et al.
Astronomy and Astrophysics. 2018. Vol. 618. P. A183-1-A183-13.
The faculty trains developers and researchers. The programme has been created based on the experience of leading American and European universities, such as Stanford University (U.S.) and EPFL (Switzerland). Also taken into consideration when creating the faculty was the School of Data Analysis, which is one of the strongest postgraduate schools in the field of computer science in Russia. The wide range of elective courses will allow each student to create his or her own educational path. In the faculty, learning is based on practice and projects.
Faculty of Computer Science will host a series of public lectures on February 20 and 26.
17.00 – 17.55. Thibaut Le Gouic (Ecole Centrale de Marseille)
Barycenters in Wasserstein spaces
The Wasserstein barycenter of a set of probabilitycommunity. This talk will introduce some of the problems in machine learning in which this tool appear as a solution. Then, after a formal definition of this notion, we will state its basics properties and study its stability with respect to the set of probability measures of which the barycenter is taken.
19.00 – 19.55. Alexey Naumov (Skoltech/CS HSE)
Large ball probabilities with applications in statistical inference
Many statistical inference problems require estimating the Kolmogorov distance between probabilities of two Gaussian elements to hit a ball in a Hilbert space. In this talk, we derive the bounds which are dimensional-free and depend on the nuclear (Schatten-one) norm of the difference between the covariance operators of the elements. We are also interested in the anti-concentration bound for a squared norm of a non-centered Gaussian element in a Hilbert space. All bounds are sharp and cannot be improved in general. A list of corresponding motivation examples and applications in statistical inference will be provided as well. The talk is based on the recent results arXiv:1708.08663 and arXiv:1703.00871.
14.00 – 14.55. Navid Talebanfard (Czech Academy of Sciences)
Hard Instances for SAT algorithms
The Boolean satisfiability problem (SAT) is the problem of deciding whether there exists a truth assignment to a given formula which makes it true. Many non-trivial exponential time algorithms are known. However the precise complexity of SAT remains elusive. Exponential Time Hypothesis (ETH) and Strong ETH are formulated to describe the possible complexities. In this talk I will review some constructions showing that several known classes of algorithms cannot refute Strong ETH. I will then present several open questions in this direction.
15.00 – 15.55. Laurent Beaudou (Université Clermont Auvergne)
Betweenness, Convexity and LatticesIn this lecture, we try to follow a natural progression from the notion of betweenness to that of convexity. Doing so, we explore some de Bruijn-Erdos problems around lines in convexity spaces. We also deal with the use of convex subsets in the framework of lattices, and discuss a potential way to extend Day’s duplication theorem to non-convex sets. This lecture is intended for a broad audience and I will be mostly interested in introducing concepts and open questions around these topics.