English
Related papers

Related papers: Shift-Curvature, SGD, and Generalization

200 papers

Recent studies have demonstrated that noise in stochastic gradient descent (SGD) is closely related to generalization: A larger SGD noise, if not too large, results in better generalization. Since the covariance of the SGD noise is…

Machine Learning · Computer Science 2020-09-29 Takashi Mori , Masahito Ueda

Although SGD with random reshuffle has been widely-used in machine learning applications, there is a limited understanding of how model characteristics affect the convergence of the algorithm. In this work, we introduce model incoherence to…

Optimization and Control · Mathematics 2020-07-08 Shaocong Ma , Yi Zhou

In this work, we reveal a strong implicit bias of stochastic gradient descent (SGD) that drives overly expressive networks to much simpler subnetworks, thereby dramatically reducing the number of independent parameters, and improving…

Machine Learning · Computer Science 2024-05-30 Feng Chen , Daniel Kunin , Atsushi Yamamura , Surya Ganguli

Stochastic Gradient Descent (SGD) has become a cornerstone of neural network optimization due to its computational efficiency and generalization capabilities. However, the gradient noise introduced by SGD is often assumed to be uncorrelated…

Machine Learning · Computer Science 2025-12-23 Marcel Kühn , Bernd Rosenow

Local SGD is a communication-efficient variant of SGD for large-scale training, where multiple GPUs perform SGD independently and average the model parameters periodically. It has been recently observed that Local SGD can not only achieve…

Machine Learning · Computer Science 2023-03-10 Xinran Gu , Kaifeng Lyu , Longbo Huang , Sanjeev Arora

The generalization performance of a machine learning algorithm such as a neural network depends in a non-trivial way on the structure of the data distribution. To analyze the influence of data structure on test loss dynamics, we study an…

Machine Learning · Statistics 2022-03-16 Blake Bordelon , Cengiz Pehlevan

Stochastic Gradient Descent (SGD) often slows in the late stage of training due to anisotropic curvature and gradient noise. We analyze preconditioned SGD in the geometry induced by a symmetric positive definite matrix $\mathbf{M}$,…

Numerical Analysis · Mathematics 2025-11-26 Mitchell Scott , Tianshi Xu , Ziyuan Tang , Alexandra Pichette-Emmons , Qiang Ye , Yousef Saad , Yuanzhe Xi

In this paper, we propose and analyze SQuARM-SGD, a communication-efficient algorithm for decentralized training of large-scale machine learning models over a network. In SQuARM-SGD, each node performs a fixed number of local SGD steps…

Machine Learning · Computer Science 2021-10-12 Navjot Singh , Deepesh Data , Jemin George , Suhas Diggavi

Recent empirical work on stochastic gradient descent (SGD) applied to over-parameterized deep learning has shown that most gradient components over epochs are quite small. Inspired by such observations, we rigorously study properties of…

Machine Learning · Computer Science 2021-10-19 Yingxue Zhou , Xinyan Li , Arindam Banerjee

The speed at which one can minimize an expected loss using stochastic methods depends on two properties: the curvature of the loss and the variance of the gradients. While most previous works focus on one or the other of these properties,…

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

Stochastic Gradient Descent (SGD) based training of neural networks with a large learning rate or a small batch-size typically ends in well-generalizing, flat regions of the weight space, as indicated by small eigenvalues of the Hessian of…

Machine Learning · Statistics 2019-12-24 Stanisław Jastrzębski , Zachary Kenton , Nicolas Ballas , Asja Fischer , Yoshua Bengio , Amos Storkey

Previous work has examined the ability of larger capacity neural networks to generalize better than smaller ones, even without explicit regularizers, by analyzing gradient based algorithms such as GD and SGD. The presence of noise and its…

Machine Learning · Computer Science 2020-05-27 Arushi Gupta

The use of low-bit quantization has emerged as an indispensable technique for enabling the efficient training of large-scale models. Despite its widespread empirical success, a rigorous theoretical understanding of its impact on learning…

Machine Learning · Statistics 2026-02-17 Dechen Zhang , Junwei Su , Difan Zou

For nonconvex objective functions, including those found in training deep neural networks, stochastic gradient descent (SGD) with momentum is said to converge faster and have better generalizability than SGD without momentum. In particular,…

Machine Learning · Computer Science 2025-07-03 Naoki Sato , Hideaki Iiduka

In this paper, we first present an explanation regarding the common occurrence of spikes in the training loss when neural networks are trained with stochastic gradient descent (SGD). We provide evidence that the spikes in the training loss…

Machine Learning · Computer Science 2024-06-07 Libin Zhu , Chaoyue Liu , Adityanarayanan Radhakrishnan , Mikhail Belkin

Momentum is known to accelerate the convergence of gradient descent in strongly convex settings without stochastic gradient noise. In stochastic optimization, such as training neural networks, folklore suggests that momentum may help deep…

Machine Learning · Computer Science 2024-04-17 Runzhe Wang , Sadhika Malladi , Tianhao Wang , Kaifeng Lyu , Zhiyuan Li

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 gradient noise of SGD is considered to play a central role in the observed strong generalization abilities of deep learning. While past studies confirm that the magnitude and the covariance structure of gradient noise are critical for…

Machine Learning · Computer Science 2020-06-22 Jingfeng Wu , Wenqing Hu , Haoyi Xiong , Jun Huan , Vladimir Braverman , Zhanxing Zhu

We study trade-offs between the population risk curvature, geometry of the noise, and preconditioning on the generalisation ability of the multipass Preconditioned Stochastic Gradient Descent (PSGD). Many practical optimisation heuristics…

Machine Learning · Computer Science 2026-03-13 Simon Vary , Tyler Farghly , Ilja Kuzborskij , Patrick Rebeschini
‹ Prev 1 2 3 10 Next ›