works on

From the 1 of 17 linked papers with an AI index.

activity
20242026
collaborators
Showing 2025Show all

7 papers · 1 filter

cs.CL2025

Initialization using Update Approximation is a Silver Bullet for Extremely Efficient Low-Rank Fine-Tuning

Kaustubh Ponkshe, Raghav Singhal, Eduard Gorbunov +3

Low-rank adapters have become standard for efficiently fine-tuning large language models, but they often fall short of achieving the performance of full fine-tuning. We propose a m…

cs.LG2025

Differentially Private Clipped-SGD: High-Probability Convergence with Arbitrary Clipping Level

Saleh Vatan Khah, Savelii Chezhegov, Shahrokh Farahmand +2

Gradient clipping is a fundamental tool in Deep Learning, improving the high-probability convergence of stochastic first-order methods like SGD, AdaGrad, and Adam under heavy-taile…

math.OC2025

Convergence of Clipped-SGD for Convex -Smooth Optimization with Heavy-Tailed Noise

Savelii Chezhegov, Aleksandr Beznosikov, Samuel Horváth +1

Gradient clipping is a widely used technique in Machine Learning and Deep Learning (DL), known for its effectiveness in mitigating the impact of heavy-tailed noise, which frequentl…

cs.LG2025

Clipping Improves Adam-Norm and AdaGrad-Norm when the Noise Is Heavy-Tailed

Savelii Chezhegov, Yaroslav Klyukin, Andrei Semenov +5

Methods with adaptive stepsizes, such as AdaGrad and Adam, are essential for training modern Deep Learning models, especially Large Language Models. Typically, the noise in the sto…

cs.LG2025

EF21 with Bells & Whistles: Six Algorithmic Extensions of Modern Error Feedback

Ilyas Fatkhullin, Igor Sokolov, Eduard Gorbunov +2

First proposed by Seide (2014) as a heuristic, error feedback (EF) is a very popular mechanism for enforcing convergence of distributed gradient-based optimization methods enhanced…

math.OC2025

Methods with Local Steps and Random Reshuffling for Generally Smooth Non-Convex Federated Optimization

Yury Demidovich, Petr Ostroukhov, Grigory Malinovsky +4

Non-convex Machine Learning problems typically do not adhere to the standard smoothness assumption. Based on empirical findings, Zhang et al. (2020b) proposed a more realistic gene…