English
Related papers

Related papers: Eliminating Sharp Minima from SGD with Truncated H…

200 papers

In this paper, we provide a theoretical study of noise geometry for minibatch stochastic gradient descent (SGD), a phenomenon where noise aligns favorably with the geometry of local landscape. We propose two metrics, derived from analyzing…

Machine Learning · Computer Science 2024-02-02 Mingze Wang , Lei Wu

Gradient compression has surfaced as a key technique to address the challenge of communication efficiency in distributed learning. In distributed deep learning, however, it is observed that gradient distributions are heavy-tailed, with…

Machine Learning · Computer Science 2024-02-07 Guangfeng Yan , Tan Li , Yuanzhang Xiao , Hanxu Hou , Linqi Song

In the vanishing learning rate regime, stochastic gradient descent (SGD) is now relatively well understood. In this work, we propose to study the basic properties of SGD and its variants in the non-vanishing learning rate regime. The focus…

Machine Learning · Statistics 2021-06-14 Kangqiao Liu , Liu Ziyin , Masahito Ueda

Neural network compression has been an increasingly important subject, not only due to its practical relevance, but also due to its theoretical implications, as there is an explicit connection between compressibility and generalization…

Machine Learning · Statistics 2024-02-13 Yijun Wan , Melih Barsbey , Abdellatif Zaidi , Umut Simsekli

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-tailed noise, which is common in training large language models.…

Machine Learning · Computer Science 2025-09-30 Saleh Vatan Khah , Savelii Chezhegov , Shahrokh Farahmand , Samuel Horváth , Eduard Gorbunov

We consider the problem of high-dimensional heavy-tailed statistical estimation in the streaming setting, which is much harder than the traditional batch setting due to memory constraints. We cast this problem as stochastic convex…

Machine Learning · Statistics 2024-10-29 Aniket Das , Dheeraj Nagaraj , Soumyabrata Pal , Arun Suggala , Prateek Varshney

Heavy-tailed noise in nonconvex stochastic optimization has garnered increasing research interest, as empirical studies, including those on training attention models, suggest it is a more realistic gradient noise condition. This paper…

Optimization and Control · Mathematics 2026-04-17 Shuhua Yu , Dusan Jakovetic , Soummya Kar

Stochastic gradient descent (SGD) is of fundamental importance in deep learning. Despite its simplicity, elucidating its efficacy remains challenging. Conventionally, the success of SGD is ascribed to the stochastic gradient noise (SGN)…

Machine Learning · Computer Science 2023-02-21 Chengli Tan , Jiangshe Zhang , Junmin Liu

Non-convex optimization problems are ubiquitous in machine learning, especially in Deep Learning. While such complex problems can often be successfully optimized in practice by using stochastic gradient descent (SGD), theoretical analysis…

Machine Learning · Computer Science 2022-02-21 Harsh Vardhan , Sebastian U. Stich

Gradient clipping is a standard safeguard for training neural networks under noisy, heavy-tailed stochastic gradients; yet, most clipping rules treat all parameters as vectors and ignore the matrix structure of modern architectures. We show…

Machine Learning · Computer Science 2026-05-13 Alexander Yukhimchuk , Mladen Kolar , Martin Takáč , Sayantan Choudhury

Stochastic gradient descent with momentum (SGDm) is one of the most popular optimization algorithms in deep learning. While there is a rich theory of SGDm for convex problems, the theory is considerably less developed in the context of deep…

Machine Learning · Statistics 2020-11-05 Umut Şimşekli , Lingjiong Zhu , Yee Whye Teh , Mert Gürbüzbalaban

Large-batch stochastic gradient descent (SGD) is widely used for training in distributed deep learning because of its training-time efficiency, however, extremely large-batch SGD leads to poor generalization and easily converges to sharp…

Machine Learning · Computer Science 2019-06-27 Kosuke Haruki , Taiji Suzuki , Yohei Hamakawa , Takeshi Toda , Ryuji Sakai , Masahiro Ozawa , Mitsuhiro Kimura

Differentially Private Stochastic Gradient Descent (DPSGD) is widely utilized to preserve training data privacy in deep learning, which first clips the gradients to a predefined norm and then injects calibrated noise into the training…

Machine Learning · Computer Science 2024-05-29 Haichao Sha , Yang Cao , Yong Liu , Yuncheng Wu , Ruixuan Liu , Hong Chen

Stochastic gradient descent (SGD) is widely believed to perform implicit regularization when used to train deep neural networks, but the precise manner in which this occurs has thus far been elusive. We prove that SGD minimizes an average…

Machine Learning · Computer Science 2018-01-17 Pratik Chaudhari , Stefano Soatto

Deep neural networks with remarkably strong generalization performances are usually over-parameterized. Despite explicit regularization strategies are used for practitioners to avoid over-fitting, the impacts are often small. Some…

Computation and Language · Computer Science 2018-11-05 Deren Lei , Zichen Sun , Yijun Xiao , William Yang Wang

It is often observed that stochastic gradient descent (SGD) and its variants implicitly select a solution with good generalization performance; such implicit bias is often characterized in terms of the sharpness of the minima. Kleinberg et…

Machine Learning · Statistics 2024-05-28 Atsushi Nitanda , Ryuhei Kikuchi , Shugo Maeda , Denny Wu

The empirical evidence indicates that stochastic optimization with heavy-tailed gradient noise is more appropriate to characterize the training of machine learning models than that with standard bounded gradient variance noise. Most…

Machine Learning · Computer Science 2026-01-28 Hongxu Chen , Ke Wei , Xiaoming Yuan , Luo Luo

We study the convergence properties and escape dynamics of Stochastic Gradient Descent (SGD) in one-dimensional landscapes, separately considering infinite- and finite-variance noise. Our main focus is to identify the time scales on which…

Machine Learning · Computer Science 2026-03-05 Dmitry Dudukalov , Artem Logachov , Vladimir Lotov , Timofei Prasolov , Evgeny Prokopenko , Anton Tarasenko

We develop a worst-case complexity theory for stochastically preconditioned stochastic gradient descent (SPSGD) and its accelerated variants under heavy-tailed noise, a setting that encompasses widely used adaptive methods such as Adam,…

Machine Learning · Computer Science 2026-02-17 Yuchen Fang , James Demmel , Javad Lavaei

The stochastic gradient descent (SGD) method and its variants are algorithms of choice for many Deep Learning tasks. These methods operate in a small-batch regime wherein a fraction of the training data, say $32$-$512$ data points, is…

Machine Learning · Computer Science 2017-02-13 Nitish Shirish Keskar , Dheevatsa Mudigere , Jorge Nocedal , Mikhail Smelyanskiy , Ping Tak Peter Tang