English
Related papers

Related papers: Why (and When) does Local SGD Generalize Better th…

200 papers

Stochastic Gradient Descent (SGD) is the most popular algorithm for training deep neural networks (DNNs). As larger networks and datasets cause longer training times, training on distributed systems is common and distributed SGD variants,…

Machine Learning · Computer Science 2019-06-17 Kwangmin Yu , Thomas Flynn , Shinjae Yoo , Nicholas D'Imperio

A number of competing hypotheses have been proposed to explain why small-batch Stochastic Gradient Descent (SGD)leads to improved generalization over the full-batch regime, with recent work crediting the implicit regularization of various…

Machine Learning · Computer Science 2022-11-30 Zachary Novack , Simran Kaur , Tanya Marwah , Saurabh Garg , Zachary C. Lipton

Intensive communication and synchronization cost for gradients and parameters is the well-known bottleneck of distributed deep learning training. Based on the observations that Synchronous SGD (SSGD) obtains good convergence accuracy while…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-04-12 Yemao Xu , Dezun Dong , Yawei Zhao , Weixia Xu , Xiangke Liao

Neural networks are usually trained by some form of stochastic gradient descent (SGD)). A number of strategies are in common use intended to improve SGD optimization, such as learning rate schedules, momentum, and batching. These are…

Neural and Evolutionary Computing · Computer Science 2015-08-13 Thomas M. Breuel

Asynchronous stochastic gradient descent (SGD) is attractive from a speed perspective because workers do not wait for synchronization. However, the Transformer model converges poorly with asynchronous SGD, resulting in substantially lower…

Computation and Language · Computer Science 2021-11-30 Alham Fikri Aji , Kenneth Heafield

This paper investigates scaling laws for local SGD in LLM training, a distributed optimization algorithm that facilitates training on loosely connected devices. Through extensive experiments, we show that local SGD achieves competitive…

Computation and Language · Computer Science 2024-09-23 Qiaozhi He , Xiaomin Zhuang , Zhihua Wu

The noise in stochastic gradient descent (SGD), caused by minibatch sampling, is poorly understood despite its practical importance in deep learning. This work presents the first systematic study of the SGD noise and fluctuations close to a…

Machine Learning · Computer Science 2022-03-09 Liu Ziyin , Kangqiao Liu , Takashi Mori , Masahito Ueda

We perform an experimental study of the dynamics of Stochastic Gradient Descent (SGD) in learning deep neural networks for several real and synthetic classification tasks. We show that in the initial epochs, almost all of the performance…

Machine Learning · Computer Science 2019-05-29 Preetum Nakkiran , Gal Kaplun , Dimitris Kalimeris , Tristan Yang , Benjamin L. Edelman , Fred Zhang , Boaz Barak

LocalSGD and SCAFFOLD are widely used methods in distributed stochastic optimization, with numerous applications in machine learning, large-scale data processing, and federated learning. However, rigorously establishing their theoretical…

Optimization and Control · Mathematics 2025-02-25 Ruichen Luo , Sebastian U Stich , Samuel Horváth , Martin Takáč

We study to what extent may stochastic gradient descent (SGD) be understood as a "conventional" learning rule that achieves generalization performance by obtaining a good fit to training data. We consider the fundamental stochastic convex…

Machine Learning · Computer Science 2023-01-13 Tomer Koren , Roi Livni , Yishay Mansour , Uri Sherman

Stochastic Gradient Descent (SGD) is the key learning algorithm for many machine learning tasks. Because of its computational costs, there is a growing interest in accelerating SGD on HPC resources like GPU clusters. However, the…

Machine Learning · Computer Science 2021-01-20 Peng Jiang , Gagan Agrawal

Most parallel neural network training methods assume homogeneous computing resources. For example, synchronous data-parallel SGD suffers from significant synchronization overhead under heterogeneous workloads, often forcing practitioners to…

Machine Learning · Computer Science 2026-02-24 Jihyun Lim , Junhyuk Jo , Chanhyeok Ko , Young Min Go , Jimin Hwa , Sunwoo Lee

For infinitesimal learning rates, stochastic gradient descent (SGD) follows the path of gradient flow on the full batch loss function. However moderately large learning rates can achieve higher test accuracies, and this generalization…

Machine Learning · Computer Science 2021-01-29 Samuel L. Smith , Benoit Dherin , David G. T. Barrett , Soham De

In this paper, we consider a general stochastic optimization problem which is often at the core of supervised learning, such as deep learning and linear classification. We consider a standard stochastic gradient descent (SGD) method with a…

Machine Learning · Statistics 2018-12-27 Lam M. Nguyen , Nam H. Nguyen , Dzung T. Phan , Jayant R. Kalagnanam , Katya Scheinberg

Decentralized stochastic gradient descent (D-SGD) allows collaborative learning on massive devices simultaneously without the control of a central server. However, existing theories claim that decentralization invariably undermines…

Machine Learning · Computer Science 2023-11-10 Tongtian Zhu , Fengxiang He , Kaixuan Chen , Mingli Song , Dacheng Tao

When training neural networks, it has been widely observed that a large step size is essential in stochastic gradient descent (SGD) for obtaining superior models. However, the effect of large step sizes on the success of SGD is not well…

Machine Learning · Computer Science 2023-02-17 Amirkeivan Mohtashami , Martin Jaggi , Sebastian Stich

In federated distributed learning, the goal is to optimize a global training objective defined over distributed devices, where the data shard at each device is sampled from a possibly different distribution (a.k.a., heterogeneous or non…

Machine Learning · Computer Science 2019-12-10 Farzin Haddadpour , Mehrdad Mahdavi

Stochastic gradient descent (SGD) has been widely used in machine learning due to its computational efficiency and favorable generalization properties. Recently, it has been empirically demonstrated that the gradient noise in several deep…

Machine Learning · Statistics 2019-06-24 Thanh Huy Nguyen , Umut Şimşekli , Mert Gürbüzbalaban , Gaël Richard

Stochastic gradient descent (SGD) exhibits strong algorithmic regularization effects in practice and plays an important role in the generalization of modern machine learning. However, prior research has revealed instances where the…

Machine Learning · Computer Science 2024-05-28 Junwei Su , Difan Zou , Chuan Wu

This work studies the generalization error of gradient methods. More specifically, we focus on how training steps $T$ and step-size $\eta$ might affect generalization in smooth stochastic convex optimization (SCO) problems. We first provide…

Machine Learning · Computer Science 2023-05-11 Peiyuan Zhang , Jiaye Teng , Jingzhao Zhang