English
Related papers

Related papers: Asynchronous Local-SGD Training for Language Model…

200 papers

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

We consider distributed optimization under communication constraints for training deep learning models. We propose a new algorithm, whose parameter updates rely on two forces: a regular gradient step, and a corrective direction dictated by…

Machine Learning · Computer Science 2022-04-29 Yunfei Teng , Wenbo Gao , Francois Chalus , Anna Choromanska , Donald Goldfarb , Adrian Weller

Stochastic gradient descent (SGD) is a widely adopted iterative method for optimizing differentiable objective functions. In this paper, we propose and discuss a novel approach to scale up SGD in applications involving non-convex functions…

Machine Learning · Statistics 2022-10-07 Saad Mohamad , Hamad Alamri , Abdelhamid Bouchachia

The performance of fully synchronized distributed systems has faced a bottleneck due to the big data trend, under which asynchronous distributed systems are becoming a major popularity due to their powerful scalability. In this paper, we…

Machine Learning · Statistics 2019-10-01 Jayanth Regatti , Gaurav Tendolkar , Yi Zhou , Abhishek Gupta , Yingbin Liang

As datasets and models become increasingly large, distributed training has become a necessary component to allow deep neural networks to train in reasonable amounts of time. However, distributed training can have substantial communication…

Machine Learning · Computer Science 2021-10-18 Jose Javier Gonzalez Ortiz , Jonathan Frankle , Mike Rabbat , Ari Morcos , Nicolas Ballas

Synchronous local stochastic gradient descent (local SGD) suffers from some workers being idle and random delays due to slow and straggling workers, as it waits for the workers to complete the same amount of local updates. In this paper, to…

Machine Learning · Computer Science 2023-05-30 Feng Zhu , Jingjing Zhang , Xin Wang

We consider speeding up stochastic gradient descent (SGD) by parallelizing it across multiple workers. We assume the same data set is shared among $n$ workers, who can take SGD steps and coordinate with a central server. Unfortunately, this…

Optimization and Control · Mathematics 2020-06-05 Artin Spiridonoff , Alex Olshevsky , Ioannis Ch. Paschalidis

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

Large-scale distributed training of deep acoustic models plays an important role in today's high-performance automatic speech recognition (ASR). In this paper we investigate a variety of asynchronous decentralized distributed training…

Computation and Language · Computer Science 2021-10-22 Xiaodong Cui , Wei Zhang , Abdullah Kayi , Mingrui Liu , Ulrich Finkler , Brian Kingsbury , George Saon , David Kung

The training of modern deep learning neural network calls for large amounts of computation, which is often provided by GPUs or other specific accelerators. To scale out to achieve faster training speed, two update algorithms are mainly…

Machine Learning · Computer Science 2020-05-15 Yemao Xu , Dezun Dong , Weixia Xu , Xiangke Liao

Stochastic gradient descent (SGD) is a well known method for regression and classification tasks. However, it is an inherently sequential algorithm at each step, the processing of the current example depends on the parameters learned from…

Machine Learning · Computer Science 2017-05-24 Saeed Maleki , Madanlal Musuvathi , Todd Mytkowicz

Distributed parallel stochastic gradient descent algorithms are workhorses for large scale machine learning tasks. Among them, local stochastic gradient descent (Local SGD) has attracted significant attention due to its low communication…

Machine Learning · Computer Science 2020-12-16 Shuheng Shen , Yifei Cheng , Jingchang Liu , Linli Xu

As the size of models and datasets grows, it has become increasingly common to train models in parallel. However, existing distributed stochastic gradient descent (SGD) algorithms suffer from insufficient utilization of computational…

Machine Learning · Computer Science 2023-08-30 Xin Zhou , Ling Chen , Houming Wu

The increasing size of deep learning models has made distributed training across multiple devices essential. However, current methods such as distributed data-parallel training suffer from large communication and synchronization overheads…

Machine Learning · Computer Science 2025-02-10 Cabrel Teguemne Fokam , Khaleelulla Khan Nazeer , Lukas König , David Kappel , Anand Subramoney

Decentralized optimization has emerged as a critical paradigm for distributed learning, enabling scalable training while preserving data privacy through peer-to-peer collaboration. However, existing methods often suffer from communication…

Machine Learning · Computer Science 2026-01-06 Yijie Zhou , Shi Pu

Distributed stochastic gradient descent (SGD) has attracted considerable recent attention due to its potential for scaling computational resources, reducing training time, and helping protect user privacy in machine learning. However, the…

Machine Learning · Computer Science 2025-02-27 Siyuan Yu , Wei Chen , H. Vincent Poor

In order to extract the best possible performance from asynchronous stochastic gradient descent one must increase the mini-batch size and scale the learning rate accordingly. In order to achieve further speedup we introduce a technique that…

Computation and Language · Computer Science 2018-09-17 Nikolay Bogoychev , Marcin Junczys-Dowmunt , Kenneth Heafield , Alham Fikri Aji

The state-of-the-art deep learning algorithms rely on distributed training systems to tackle the increasing sizes of models and training data sets. Minibatch stochastic gradient descent (SGD) algorithm requires workers to halt forward/back…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-06-02 Qinggang Zhou , Yawen Zhang , Pengcheng Li , Xiaoyong Liu , Jun Yang , Runsheng Wang , Ru Huang

Asynchronous stochastic gradient descent (SGD) enables scalable distributed training but suffers from gradient staleness. Existing mitigation strategies, such as delay-adaptive learning rates and staleness-aware filtering, typically…

Machine Learning · Computer Science 2026-05-15 Tehila Dahan , Roie Reshef , Sharon Goldstein , Kfir Y. Levy

Recently, the technique of local updates is a powerful tool in centralized settings to improve communication efficiency via periodical communication. For decentralized settings, it is still unclear how to efficiently combine local updates…

Machine Learning · Statistics 2021-04-06 Xiang Li , Wenhao Yang , Shusen Wang , Zhihua Zhang