Andrii Semenov
Andrii Semenov
Home
Education
Work Experience
Posts
Publications
Talks
Projects
Light
Dark
Automatic
Posts
Visiting Martin Takáč at MBZUAI
Between 15 February and 15 April, I will work as a Visiting Student in the Machine Learning Department of MBZUAI under the supervision of Professor Martin Takáč.
Feb 14, 2025
0 min read
New paper out
“Sign Operator for Coping with Heavy-Tailed Noise: High Probability Convergence Bounds with Extensions to Distributed Optimization and Comparison Oracle” – joint work with Nikita Kornilov, Philip Zmushko, Alexander Gasnikov and Aleksandr Beznosikov.
Feb 11, 2025
1 min read
PDF
arXiv
New paper out
“Just a Simple Transformation is Enough for Data Protection in Vertical Federated Learning” – joint work with Philip Zmushko, Alexander Pichugin and Aleksandr Beznosikov.
Dec 16, 2024
1 min read
PDF
arXiv
Talk on the Defense against Feature Reconstruction attacks
I gave a talk on the Defense against Feature Reconstruction attacks at
MLO Group Meeting
.
Oct 9, 2024
0 min read
Slides
I was awarded 2 scholarships from MIPT
This Autumn I was awarded 2 scholarships. K. V. Rudakov scientific academic scholarship and 1st degree personal scholarship for contributions to the development of numerical optimization methods. Overall cost is approximately $3300 during one semeter.
Sep 30, 2024
0 min read
ICOMP 1 accepted paper (with Oral Presentation)
“Bregman Proximal Method for Efficient Communications under Similarity” – joint work with Aleksandr Beznosikov, Darina Dvinskikh, Dmity Bylinkin, Alexander Gasnikov.
Sep 16, 2024
1 min read
PDF
Poster
arXiv
Admission to MIPT
I am accepted at MIPT to follow the Computer Science and Informatics Master’s program at the Department of Intelligent Systems.
Aug 10, 2024
0 min read
New paper out
“Mixed Newton Method for Optimization in Complex Spaces” – joint work with Nikita Yudin, Roland Hildebrand, Sergey Bakhurin, Alexander Degtyarev, Anna Lisachenko, Ilya Kuruzov, Mohammad Alkousa.
Jul 29, 2024
1 min read
PDF
arXiv
New paper out
“Gradient Clipping Improves AdaGrad when the Noise Is Heavy-Tailed” – joint work with Savelii Chezhegov, Yaroslav Klyukin, Aleksandr Beznosikov, Alexander Gasnikov, Samuel Horváth, Martin Takáč and Eduard Gorbunov.
Jun 6, 2024
1 min read
PDF
arXiv
New paper out
“Sparse Concept Bottleneck Models: Gumbel Tricks in Contrastive Learning” – joint work with Vladimir Ivanov, Aleksandr Beznosikov, Alexander Gasnikov.
Apr 4, 2024
1 min read
PDF
arXiv
»
Cite
×