English
Related papers

Related papers: Slow and Stale Gradients Can Win the Race

200 papers

Asynchronous iterative methods tolerate straggling processors by allowing workers to proceed with stale data, but at a cost: the iterates become inconsistent, potentially degrading convergence. We investigate whether convergence…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-28 Evan Coleman , Masha Sosonkina

Coded computation techniques provide robustness against straggling servers in distributed computing, with the following limitations: First, they increase decoding complexity. Second, they ignore computations carried out by straggling…

Machine Learning · Computer Science 2018-11-29 Emre Ozfatura , Sennur Ulukus , Deniz Gunduz

In this paper, we investigate a general class of stochastic gradient descent (SGD) algorithms, called Conditioned SGD, based on a preconditioning of the gradient direction. Using a discrete-time approach with martingale tools, we establish…

Statistics Theory · Mathematics 2023-10-17 Rémi Leluc , François Portier

Huge scale machine learning problems are nowadays tackled by distributed optimization algorithms, i.e. algorithms that leverage the compute power of many devices for training. The communication overhead is a key bottleneck that hinders…

Machine Learning · Computer Science 2018-11-30 Sebastian U. Stich , Jean-Baptiste Cordonnier , Martin Jaggi

Distributed asynchronous SGD has become widely used for deep learning in large-scale systems, but remains notorious for its instability when increasing the number of workers. In this work, we study the dynamics of distributed asynchronous…

Machine Learning · Computer Science 2018-05-23 Joeri Hermans , Gilles Louppe

A stochastic gradient method for synchronous distributed optimization is studied. For reducing communication cost, we particularly focus on utilization of compression of communicated gradients. Several work has shown that {\it{sparsified}}…

Optimization and Control · Mathematics 2020-06-22 Tomoya Murata , Taiji Suzuki

Modern distributed optimization methods mostly rely on traditional synchronous approaches, despite substantial recent progress in asynchronous optimization. We revisit Synchronous SGD and its robust variant, called $m$-Synchronous SGD, and…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-02-04 Grigory Begunov , Alexander Tyurin

Modern large scale machine learning applications require stochastic optimization algorithms to be implemented on distributed computational architectures. A key bottleneck is the communication overhead for exchanging information, such as…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-10-25 Zijie Yan

Background: Distributed training is essential for large scale training of deep neural networks (DNNs). The dominant methods for large scale DNN training are synchronous (e.g. All-Reduce), but these require waiting for all workers in each…

Machine Learning · Computer Science 2023-09-26 Niv Giladi , Shahar Gottlieb , Moran Shkolnik , Asaf Karnieli , Ron Banner , Elad Hoffer , Kfir Yehuda Levy , Daniel Soudry

Synchronous stochastic gradient descent (SGD) is the most common method used for distributed training of deep learning models. In this algorithm, each worker shares its local gradients with others and updates the parameters using the…

Machine Learning · Computer Science 2020-09-22 Negar Foroutan Eghlidi , Martin Jaggi

Federated Learning is a novel paradigm that involves learning from data samples distributed across a large network of clients while the data remains local. It is, however, known that federated learning is prone to multiple system challenges…

Machine Learning · Computer Science 2021-01-01 Amirhossein Reisizadeh , Isidoros Tziotis , Hamed Hassani , Aryan Mokhtari , Ramtin Pedarsani

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

Recent developments on large-scale distributed machine learning applications, e.g., deep neural networks, benefit enormously from the advances in distributed non-convex optimization techniques, e.g., distributed Stochastic Gradient Descent…

Optimization and Control · Mathematics 2019-05-13 Hao Yu , Rong Jin , Sen Yang

We study asynchronous finite sum minimization in a distributed-data setting with a central parameter server. While asynchrony is well understood in parallel settings where the data is accessible by all machines -- e.g., modifications of…

Machine Learning · Computer Science 2021-03-11 Margalit Glasgow , Mary Wootters

Stochastic Gradient Descent (SGD) is the standard numerical method used to solve the core optimization problem for the vast majority of machine learning (ML) algorithms. In the context of large scale learning, as utilized by many Big Data…

Distributed, Parallel, and Cluster Computing · Computer Science 2015-10-06 Janis Keuper , Franz-Josef Pfreundt

We consider the decentralized stochastic asynchronous optimization setup, where many workers asynchronously calculate stochastic gradients and asynchronously communicate with each other using edges in a multigraph. For both homogeneous and…

Optimization and Control · Mathematics 2024-11-05 Alexander Tyurin , Peter Richtárik

Variance reduction (VR) methods boost the performance of stochastic gradient descent (SGD) by enabling the use of larger, constant stepsizes and preserving linear convergence rates. However, current variance reduced SGD methods require…

Machine Learning · Computer Science 2017-04-10 Soham De , Gavin Taylor , Tom Goldstein

Gradient-based distributed learning in Parameter Server (PS) computing architectures is subject to random delays due to straggling worker nodes, as well as to possible communication bottlenecks between PS and workers. Solutions have been…

Information Theory · Computer Science 2020-04-09 Jingjing Zhang , Osvaldo Simeone

We propose a general framework for distributed stochastic optimization under delayed gradient models. In this setting, $n$ local agents leverage their own data and computation to assist a central server in minimizing a global objective…

Optimization and Control · Mathematics 2026-03-04 Xinran Zheng , Tara Javidi , Behrouz Touri

We provide tight finite-time convergence bounds for gradient descent and stochastic gradient descent on quadratic functions, when the gradients are delayed and reflect iterates from $\tau$ rounds ago. First, we show that without stochastic…

Optimization and Control · Mathematics 2018-06-28 Yossi Arjevani , Ohad Shamir , Nathan Srebro