Ilya Valerievich Levin

Ilya Valerievich Levin serves as a Junior Research Fellow at the International Laboratory of Stochastic Algorithms and Multidimensional Data Analysis and as a Teacher in the Department of Big Data and Information Retrieval, both within the Faculty of Computer Science at the National Research University Higher School of Economics (HSE). He began his tenure at HSE in 2020 and has accumulated 3 years of scientific and teaching experience. His primary affiliations include the Institute of Artificial Intelligence and Digital Sciences where he works under the supervision of A. A. Naumov.

• ORCID: 0000-0002-9708-8189
• Researcher ID: GZL-7944-2022
• Location: Pokrovsky Boulevard, 11, office T926
• Office Hours: 12:00-18:00

Levin's research expertise centers on reinforcement learning methodologies, stochastic approximation algorithms, and multidimensional data analysis. His work bridges theoretical foundations with practical applications in machine learning systems, particularly focusing on federated learning environments and evaluation frameworks for reinforcement learning algorithms. His research demonstrates strong mathematical foundations with applications across distributed computing systems.

His publication record shows a clear trajectory in advancing reinforcement learning techniques, with recent work presented at NeurIPS 2024 addressing heterogeneity challenges in federated learning systems. His research combines theoretical rigor with practical implementation concerns, reflecting the interdisciplinary nature of modern artificial intelligence research.

• Letter of gratitude from the Faculty of Computer Science (July 2024)
• Bonus for publication in List A journal (2025-2026)
• HSE Personnel Reserve - New Researchers category (2022)

Levin has been actively involved in teaching Probability Theory, Mathematical Statistics, and Mathematical Foundations of Reinforcement Learning across multiple academic programs including bachelor's and master's degrees. His teaching spans both Russian and English language instruction formats, including specialized programs like Mago-Lego. He has consistently taught core mathematical courses for computer science students since the 2022/2023 academic year, demonstrating commitment to foundational education in the field.