paper-with-me

Papers

AdaGrad-Diff: A New Version of the Adaptive Gradient Algorithm

2026-02-13 · Matia Bojovic, Saverio Salzo, Massimiliano Pontil arxiv

Vanilla gradient methods are often highly sensitive to the choice of stepsize, which typically requires manual tuning. Adaptive methods alleviate this issue and have therefore become widely used. Among them, AdaGrad has been particularly influential. In this paper, we propose an AdaGrad-style adaptive method in which the adaptation is driven by the cumulative squared norms of successive gradient differences rather than gradient norms themselves. The key idea is that when gradients vary little across iterations, the stepsize is not unnecessarily reduced, while significant gradient fluctuations, reflecting curvature or instability, lead to automatic stepsize damping. Numerical experiments demonstrate that the proposed method is more robust than AdaGrad in several practically relevant settings.

📄 PDF Abstract BibTeX arXiv:2602.13112

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Private Adaptive Gradient Methods for Convex Optimization

2021-06-25 · Hilal Asi, John Duchi, Alireza Fallah, Omid Javidbakht 외

We study adaptive methods for differentially private convex optimization, proposing and analyzing differentially private variants of a Stochastic Gradient Descent (SGD) algorithm with adaptive stepsizes, as well as the A…

Linear Convergence of Adaptive Stochastic Gradient Descent

2019-08-28 · Yuege Xie, Xiaoxia Wu, Rachel Ward

We prove that the norm version of the adaptive stochastic gradient method (AdaGrad-Norm) achieves a linear convergence rate for a subset of either strongly convex functions or non-convex functions that satisfy the Polyak…

CompAdaGrad: A Compressed, Complementary, Computationally-Efficient Adaptive Gradient Method

2016-09-12 · Nishant A. Mehta, Alistair Rendell, Anish Varghese, Christfried Webers

The adaptive gradient online learning method known as AdaGrad has seen widespread use in the machine learning community in stochastic and adversarial online learning problems and more recently in deep learning methods. T…

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

2024-06-06 · Savelii Chezhegov, Yaroslav Klyukin, Andrei Semenov, Aleksandr Beznosikov 외

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 stochastic gradients is heavy-tailed for th…

Stochastic Optimization

AdaGrad stepsizes: Sharp convergence over nonconvex landscapes

2018-06-05 · Rachel Ward, Xiaoxia Wu, Leon Bottou

Adaptive gradient methods such as AdaGrad and its variants update the stepsize in stochastic gradient descent on the fly according to the gradients received along the way; such methods have gained widespread use in large…

Stochastic Optimization