Fatkhullin, Ilyas

16 publications

JMLR 2025 EF21 with Bells & Whistles: Six Algorithmic Extensions of Modern Error Feedback Ilyas Fatkhullin, Igor Sokolov, Eduard Gorbunov, Zhize Li, Peter Richtárik
AISTATS 2025 From Gradient Clipping to Normalization for Heavy Tailed SGD Florian Hübler, Ilyas Fatkhullin, Niao He
NeurIPS 2025 Natural Gradient VI: Guarantees for Non-Conjugate Models Fangyuan Sun, Ilyas Fatkhullin, Niao He
ICML 2025 Safe-EF: Error Feedback for Non-Smooth Constrained Optimization Rustem Islamov, Yarden As, Ilyas Fatkhullin
NeurIPS 2025 Second-Order Optimization Under Heavy-Tailed Noise: Hessian Clipping and Sample Complexity Limits Abdurakhmon Sadiev, Peter Richtárik, Ilyas Fatkhullin
NeurIPSW 2024 From Gradient Clipping to Normalization for Heavy Tailed SGD Florian Hübler, Ilyas Fatkhullin, Niao He
AISTATS 2024 Taming Nonconvex Stochastic Mirror Descent with General Bregman Divergence Ilyas Fatkhullin, Niao He
NeurIPS 2023 Momentum Provably Improves Error Feedback! Ilyas Fatkhullin, Alexander Tyurin, Peter Richtarik
ICMLW 2023 Momentum Provably Improves Error Feedback! Ilyas Fatkhullin, Alexander Tyurin, Peter Richtárik
ICML 2023 Reinforcement Learning with General Utilities: Simpler Variance Reduction and Large State-Action Space Anas Barakat, Ilyas Fatkhullin, Niao He
NeurIPSW 2023 Stochastic Optimization Under Hidden Convexity Ilyas Fatkhullin, Niao He, Yifan Hu
ICML 2023 Stochastic Policy Gradient Methods: Improved Sample Complexity for Fisher-Non-Degenerate Policies Ilyas Fatkhullin, Anas Barakat, Anastasia Kireeva, Niao He
NeurIPS 2023 Two Sides of One Coin: The Limits of Untuned SGD and the Power of Adaptive Methods Junchi Yang, Xiang Li, Ilyas Fatkhullin, Niao He
ICML 2022 3PC: Three Point Compressors for Communication-Efficient Distributed Training and a Better Theory for Lazy Aggregation Peter Richtarik, Igor Sokolov, Elnur Gasanov, Ilyas Fatkhullin, Zhize Li, Eduard Gorbunov
NeurIPS 2022 Sharp Analysis of Stochastic Optimization Under Global Kurdyka-Lojasiewicz Inequality Ilyas Fatkhullin, Jalal Etesami, Niao He, Negar Kiyavash
NeurIPS 2021 EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback Peter Richtarik, Igor Sokolov, Ilyas Fatkhullin