|
January 24, 2022 / Evgeny Feigin (HSE Univ., Skoltech) = = Quadratic algebras, Koszul duality and Donaldson-Thomas invariants January 31, 2022 / Vasily Bolbachan (HSE Univ.) = = Chow dilogarithm and reciprocity laws February 7, 2022 / Yegor Zenkevich (SISSA, INFN, ITEP, ITMP MSU) = = On R-matrices and branes February 14, 2022 / Alexander Abanov (SCGP, Stony Brook Univ.) = = Quantum Hall effect and adiabatic transport (1/2) February 21, 2022 / Georgii Shuklin (Skoltech, HSE Univ.) = = Kato-Nakayama motives April 11, 2022 / Vasily Krylov (MIT, HSE Univ.) = = Symplectic duality and equivariant Hikita-Nakajima conjecture for ADHM spaces April 25, 2022 / Ruotao Yang (Skoltech) = = Untwisted Gaiotto equivalence for GL(M|N) |
Artificial intelligence has become a very hot topic in the recent years due to significant success in such tasks as image recognition and generation, text processing and others. Most of these results are experimental and empirical and lack strong mathematical foundations. On the other hand, techniques from different areas of mathematics and physics (such as algebraic geometry and topology) have been proven useful for AI.
In this talk, I will briefly introduce the basic problem formulations in machine learning such as supervised and unsupervised learning, describe who different geometrical ideas can provide insights on the theoretical analysis of the properties of the machine learning algorithms and also formulate several open problems