Rustem Takhanov

Connected with DSML. This person's Telegram account has been confirmed through DSML Kazakhstan.

@k-nic

Rustem Takhanov is a mathematician at Nazarbayev University whose research spans discrete geometry, combinatorics, optimization, machine learning theory, and theoretical computer science.

Astana, KazakhstanNazarbayev University
View full research profile

About

His recent work focuses on high-dimensional kissing arrangements and sphere packings. In 2026, together with collaborators, he constructed a configuration of 841 spheres kissing a central sphere in 12-dimensional Euclidean space, improving a lower bound that had stood since the work of Leech and Sloane in 1971.

Research and publications

View full research profile

My research focuses on the interplay between machine learning, optimization, and mathematics. I develop algorithmic and computational methods for solving challenging problems in machine learning theory, discrete mathematics, and theoretical computer science. A recurring theme of my work is the use of large-scale optimization and high-performance computing as tools for scientific discovery. My contributions span machine learning theory, kernel methods, natural language processing, graph algorithms, combinatorial optimization, and high-dimensional geometry. In recent years, I have become particularly interested in computational approaches to mathematical research, where modern optimization techniques and GPU-based computation can uncover structures that are difficult to access through traditional analytical methods alone. More broadly, I aim to build bridges between theoretical foundations and practical computation, developing methods that are both mathematically rigorous and computationally effective.

Machine learning theoryoptimizationtheoretical computer sciencediscrete mathematicshigh-dimensional geometry