publications

in reverse chronological order; * indicates equal contribution.

2026

  1. CPAL
    Byzantine-Robust Optimization under (L_0, L_1)-Smoothness
    Arman Bolatov, Samuel Horváth, Martin Takáč, and 1 more author
    In Conference on Parsimony and Learning (CPAL), Proceedings Track, 2026
  2. Where Does Warm-Up Come From? Adaptive Scheduling for Norm-Constrained Optimizers
    Andrey Riabinin, Andrey Veprikov, Arman Bolatov, and 2 more authors
    arXiv preprint arXiv:2602.05813, 2026

2025

  1. Who to Trust? Aggregating Client Knowledge in Logit-Based Federated Learning
    Viktor Kovalchuk, Daniil Son, Arman Bolatov, and 6 more authors
    arXiv preprint arXiv:2509.15147, 2025
  2. Preconditioned Norms: A Unified Framework for Steepest Descent, Quasi-Newton and Adaptive Methods
    Andrey Veprikov, Arman Bolatov, Samuel Horváth, and 3 more authors
    arXiv preprint arXiv:2510.10777, 2025
  3. Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute Scaling
    Ivan Rodkin, Daniil Orel, Konstantin Smirnov, and 2 more authors
    arXiv preprint arXiv:2508.16745, 2025
  4. Overspecified Mixture Discriminant Analysis: Exponential Convergence, Statistical Guarantees, and Remote Sensing Applications
    Arman Bolatov, Adam Legg, Igor Melnykov, and 3 more authors
    arXiv preprint arXiv:2510.27056, 2025
  5. Micromachines
    Fourier Approximation of magMEMS Oscillations: Neural Network Space Handling
    Piotr Skrzypacz, Arman Bolatov, Adam Dziedzic, and 4 more authors
    Micromachines, 2025
  6. SAM
    Classifier Performance on Long-Tail Distributions
    Alisher Pak, Maxat Tezekbayev, Arman Bolatov, and 3 more authors
    Statistical Analysis and Data Mining: An ASA Data Science Journal, 2025
  7. Stat. Papers
    Convergence of the EM Algorithm in KL Distance for Overspecified Gaussian Mixtures
    Adam Legg, Igor Melnykov, Arman Bolatov, and 2 more authors
    Statistical Papers, 2025
  8. Optimizing dynamic pull-in threshold and periodic trajectories for magnetically actuated MEMS (magMEMS) in wearable sensors
    Lin Zhao, Ji-Huan He, Piotr Skrzypacz, and 2 more authors
    Frontiers in Physics, 2025
  9. MLJ
    Gradient Descent Fails to Learn High-frequency Functions and Modular Arithmetic
    Rustem Takhanov, Maxat Tezekbayev, Alisher Pak, and 2 more authors
    Machine Learning, 2025

2024

  1. ACML
    Intractability of Learning the Discrete Logarithm with Gradient-Based Methods
    Rustem Takhanov, Maxat Tezekbayev, Alisher Pak, and 3 more authors
    In Asian Conference on Machine Learning (ACML), 2024

2023

  1. Nutrients
    A Central Asian Food Dataset for Personalized Dietary Interventions
    Aknur Karabay, Arman Bolatov, Huseyin Atakan Varol, and 1 more author
    Nutrients, 2023
  2. ECAI
    Long Tail Theory Under Gaussian Mixtures
    Arman Bolatov, Maxat Tezekbayev, Igor Melnykov, and 3 more authors
    In European Conference on Artificial Intelligence (ECAI), 2023
  3. Application of Image Processing in Evaluation of Hydraulic Fracturing with Liquid Nitrogen: A Case Study of Coal Samples from Karaganda Basin
    Sotirios N. Longinos, Arman Bolatov, Piotr Skrzypacz, and 1 more author
    Applied Sciences, 2023
  4. Empirical Analysis of the AdaBoost’s Error Bound
    Arman Bolatov and Kuat Dauletbek
    arXiv preprint arXiv:2302.00880, 2023