We cordially invite you to attend the KSE Mathematics Seminar on the topic: “Random Matrix Theory and Machine Learning”

Speaker: Ievgenii Afanasiev (Kyiv School of Economics).

Ievgenii is a Simons Postdoctoral Fellow in Mathematics at KSE. He received his PhD in 2021 from the B.I. Verkin Institute for Low Temperature Physics and Engineering of the NAS of Ukraine in Kharkiv. His research focuses on random matrix theory and its applications, particularly in machine learning theory.This talk explores the connections between Random Matrix Theory (RMT) and Machine Learning. In the first part, I will discuss the historical development of RMT, starting with Eugene Wigner's work on the statistical properties of energy levels in complex quantum systems. Since then, RMT has found applications in diverse fields: from physics to financial markets and neurobiology. I will introduce key examples of random matrices and discuss the concept of universality - where large matrices exhibit the same statistical properties, regardless of the specific probabilistic distribution of their entries.

The second part of the talk will bridge RMT with deep neural networks (DNNs), a powerful tool in Machine Learning. I will explain what DNNs are from a mathematical perspective and how RMT can be used to improve DNN performance.

The seminar will take place on November 19 from 16:30 to 17:50 at the KSE Feofania campus, located at 6/11 Miroshnychenko Street, Room 2.The event will be held in English.

We warmly invite students, researchers, and anyone interested in the intersection of mathematics and machine learning to attend. Come explore cutting-edge ideas, engage in discussion, and become part of the growing KSE research community.