Rogozin Alexander Viktorovich
Receives an MIPT award for the contribution to numerical methods since 2020
Education
Graduated from Yandex School for Data Analysis in 2019.
Graduated from MIPT, Phystech school of applied mathematics and informatics in 2020.
PhD student at Phystech school of applied mathematics and informatics since 2020.
Work Experience
2019 - 2021 — teaching assistant at MIPT
2019 - 2022 — MIPT-Huawei, digital pre-distortion (DPD) joint projects. Writing code for models, experiments, and compiling reports. Subject area: digital signal preprocessing.
2020 - 2022 — junior researcher at MIPT
2020 - 2022 — leading researcher of joint MIPT--Huawei project on signal processing
2023 - 2024 — A joint project of MIPT-Huawei, computing force networks, Writing code for models, optimization methods, literature analysis, and reporting. Responsible for the project. Subject area: flows in telecom networks, network design.
Since 2022 — researcher at MIPT
Teaching
MIPT, Probability theory, 2019 - 2021
Research Interests
1. Decentralized optimization.
2. Traffic flows
3. Average-case analysis, but there is only one publication.
Research Projects and Grants
Scientific work:
different tasks in decentralized optimization. Cross—constraint optimization (together with Demyan Yarmoshik et al.) - one article on ICLR 2025, the second submitted on ICML 2026. Several works on MOTOR/OPTIMA, also on distributed optimization (dual smoothing, consensus procedure for different algorithms). I am submitting another paper (Newton with a consensus procedure) to IEEE Access (together with Artem Agafonov, Anton Novitsky, etc.).
Practical projects:
in 2024. — a joint project with Huawei on telecom networks. Calculation of network flows, restoration of the correspondence matrix, network design, optimization of large networks. The project duration is 1.5 years (actually almost 2 years). The most interesting result is a dual simplex method with a restart for fast processing of added constraints (the algorithm was proposed by Roland).
Публикации
Aleksandr Beznosikov, Georgiy Kormakov, Alexander Grigorievskiy, Mikhail Rudakov, Ruslan Nazykov, Alexander Rogozin, Anton Vakhrushev, Andrey Savchenko, Martin Takáč, Alexander Gasnikov. Exploring New Frontiers in Vertical Federated Learning: the Role of Saddle Point Reformulation (2026) Journal of Optimization Theory and Applications DOI Q1
2025
Demyan Yarmoshik, Alexander Rogozin, Nikita Kiselev, Daniil Dorin, Alexander Gasnikov, Dmitry Kovalev. Decentralized Optimization with Coupled Constraints (2025) The International Conference on Learning Representations (ICLR) A*
2024
2023
Demyan Yarmoshik, Alexander Rogozin, Alexander Gasnikov. Decentralized optimization with affine constraints over time-varying networks (2023) Scopus DOI Q3
2022
Yarmoshik D., Rogozin A., Khamisov O.O., Dvurechensky P., Gasnikov A. Decentralized Convex Optimization Under Affine Constraints for Power Systems Control // Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 13367, P. 62 - 75 Scopus WOS DOI Q3
2021
Rogozin, A., Lukoshkin, V., Gasnikov, A., Kovalev, D., Shulgin, E. Towards Accelerated Rates for Distributed Optimization over Time-Varying Networks (2021) Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 13078 LNCS, pp. 258-272. Scopus WOS DOI Q3
Beznosikov, A., Rogozin, A., Kovalev, D., Gasnikov, A. Near-Optimal Decentralized Algorithms for Saddle Point Problems over Time-Varying Networks (2021) Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 13078 LNCS, pp. 246-257. Scopus WOS DOI Q4
2020
Dmitry Metelev, Savelii Chezhegov, Alexander Rogozin, Aleksandr Beznosikov, Alexander Sholokhov, Alexander Gasnikov, Dmitry Kovalev. Decentralized finite-sum optimization over time-varying networks (2024)