Bernoulli-LoRA: A Theoretical Framework for Randomized Low-Rank Adaptation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Sokolov, Igor, Sadiev, Abdurakhmon, Demidovich, Yury, Al-Qahtani, Fawaz S, Richtárik, Peter
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911093607104512
author Sokolov, Igor
Sadiev, Abdurakhmon
Demidovich, Yury
Al-Qahtani, Fawaz S
Richtárik, Peter
author_facet Sokolov, Igor
Sadiev, Abdurakhmon
Demidovich, Yury
Al-Qahtani, Fawaz S
Richtárik, Peter
contents Parameter-efficient fine-tuning (PEFT) has emerged as a crucial approach for adapting large foundational models to specific tasks, particularly as model sizes continue to grow exponentially. Among PEFT methods, Low-Rank Adaptation (LoRA) (arXiv:2106.09685) stands out for its effectiveness and simplicity, expressing adaptations as a product of two low-rank matrices. While extensive empirical studies demonstrate LoRA's practical utility, theoretical understanding of such methods remains limited. Recent work on RAC-LoRA (arXiv:2410.08305) took initial steps toward rigorous analysis. In this work, we introduce Bernoulli-LoRA, a novel theoretical framework that unifies and extends existing LoRA approaches. Our method introduces a probabilistic Bernoulli mechanism for selecting which matrix to update. This approach encompasses and generalizes various existing update strategies while maintaining theoretical tractability. Under standard assumptions from non-convex optimization literature, we analyze several variants of our framework: Bernoulli-LoRA-GD, Bernoulli-LoRA-SGD, Bernoulli-LoRA-PAGE, Bernoulli-LoRA-MVR, Bernoulli-LoRA-QGD, Bernoulli-LoRA-MARINA, and Bernoulli-LoRA-EF21, establishing convergence guarantees for each variant. Additionally, we extend our analysis to convex non-smooth functions, providing convergence rates for both constant and adaptive (Polyak-type) stepsizes. Through extensive experiments on various tasks, we validate our theoretical findings and demonstrate the practical efficacy of our approach. This work is a step toward developing theoretically grounded yet practically effective PEFT methods.
format Preprint
id arxiv_https___arxiv_org_abs_2508_03820
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bernoulli-LoRA: A Theoretical Framework for Randomized Low-Rank Adaptation
Sokolov, Igor
Sadiev, Abdurakhmon
Demidovich, Yury
Al-Qahtani, Fawaz S
Richtárik, Peter
Machine Learning
Optimization and Control
Parameter-efficient fine-tuning (PEFT) has emerged as a crucial approach for adapting large foundational models to specific tasks, particularly as model sizes continue to grow exponentially. Among PEFT methods, Low-Rank Adaptation (LoRA) (arXiv:2106.09685) stands out for its effectiveness and simplicity, expressing adaptations as a product of two low-rank matrices. While extensive empirical studies demonstrate LoRA's practical utility, theoretical understanding of such methods remains limited. Recent work on RAC-LoRA (arXiv:2410.08305) took initial steps toward rigorous analysis. In this work, we introduce Bernoulli-LoRA, a novel theoretical framework that unifies and extends existing LoRA approaches. Our method introduces a probabilistic Bernoulli mechanism for selecting which matrix to update. This approach encompasses and generalizes various existing update strategies while maintaining theoretical tractability. Under standard assumptions from non-convex optimization literature, we analyze several variants of our framework: Bernoulli-LoRA-GD, Bernoulli-LoRA-SGD, Bernoulli-LoRA-PAGE, Bernoulli-LoRA-MVR, Bernoulli-LoRA-QGD, Bernoulli-LoRA-MARINA, and Bernoulli-LoRA-EF21, establishing convergence guarantees for each variant. Additionally, we extend our analysis to convex non-smooth functions, providing convergence rates for both constant and adaptive (Polyak-type) stepsizes. Through extensive experiments on various tasks, we validate our theoretical findings and demonstrate the practical efficacy of our approach. This work is a step toward developing theoretically grounded yet practically effective PEFT methods.
title Bernoulli-LoRA: A Theoretical Framework for Randomized Low-Rank Adaptation
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2508.03820