English
Related papers

Related papers: Distributed learning with compressed gradients

200 papers

We propose an algorithm for distributed optimization over time-varying communication networks. Our algorithm uses an optimized ratio between the number of rounds of communication and gradient evaluations to achieve fast convergence. The…

Optimization and Control · Mathematics 2020-01-08 Bryan Van Scoy , Laurent Lessard

Asynchronous distributed algorithms are a popular way to reduce synchronization costs in large-scale optimization, and in particular for neural network training. However, for nonsmooth and nonconvex objectives, few convergence guarantees…

Optimization and Control · Mathematics 2020-07-14 Vyacheslav Kungurtsev , Malcolm Egan , Bapi Chatterjee , Dan Alistarh

A lot of effort has been invested into characterizing the convergence rates of gradient based algorithms for non-linear convex optimization. Recently, motivated by large datasets and problems in machine learning, the interest has shifted…

Distributed, Parallel, and Cluster Computing · Computer Science 2012-07-23 Konstantinos I. Tsianos , Michael G. Rabbat

We develop and analyze an asynchronous algorithm for distributed convex optimization when the objective writes a sum of smooth functions, local to each worker, and a non-smooth function. Unlike many existing methods, our distributed…

Optimization and Control · Mathematics 2019-12-13 Konstantin Mishchenko , Franck Iutzeler , Jérôme Malick

Emerging applications of machine learning in numerous areas involve continuous gathering of and learning from streams of data. Real-time incorporation of streaming data into the learned models is essential for improved inference in these…

Machine Learning · Computer Science 2020-12-01 Matthew Nokleby , Haroon Raja , Waheed U. Bajwa

One of the most widely used methods for solving large-scale stochastic optimization problems is distributed asynchronous stochastic gradient descent (DASGD), a family of algorithms that result from parallelizing stochastic gradient descent…

Optimization and Control · Mathematics 2021-07-08 Zhengyuan Zhou , Panayotis Mertikopoulos , Nicholas Bambos , Peter W. Glynn , Yinyu Ye

The paper considers distributed stochastic optimization over randomly switching networks, where agents collaboratively minimize the average of all agents' local expectation-valued convex cost functions. Due to the stochasticity in gradient…

Optimization and Control · Mathematics 2022-04-07 Jinlong Lei , Peng Yi , Jie Chen , Yiguang Hong

Training large machine learning models requires a distributed computing approach, with communication of the model updates being the bottleneck. For this reason, several methods based on the compression (e.g., sparsification and/or…

Machine Learning · Computer Science 2023-12-29 Konstantin Mishchenko , Eduard Gorbunov , Martin Takáč , Peter Richtárik

With the rapid increase of big data, distributed Machine Learning (ML) has been widely applied in training large-scale models. Stochastic Gradient Descent (SGD) is arguably the workhorse algorithm of ML. Distributed ML models trained by SGD…

Machine Learning · Computer Science 2021-12-09 Keyu Yang , Lu Chen , Zhihao Zeng , Yunjun Gao

Although the distributed machine learning methods can speed up the training of large deep neural networks, the communication cost has become the non-negligible bottleneck to constrain the performance. To address this challenge, the gradient…

Machine Learning · Computer Science 2022-01-25 An Xu , Zhouyuan Huo , Heng Huang

Distributed training of massive machine learning models, in particular deep neural networks, via Stochastic Gradient Descent (SGD) is becoming commonplace. Several families of communication-reduction methods, such as quantization,…

Machine Learning · Computer Science 2018-09-28 Dan Alistarh , Torsten Hoefler , Mikael Johansson , Sarit Khirirat , Nikola Konstantinov , Cédric Renggli

Large-scale distributed training of Deep Neural Networks (DNNs) on state-of-the-art platforms is expected to be severely communication constrained. To overcome this limitation, numerous gradient compression techniques have been proposed and…

In distributed training of deep neural networks, people usually run Stochastic Gradient Descent (SGD) or its variants on each machine and communicate with other machines periodically. However, SGD might converge slowly in training some deep…

Machine Learning · Computer Science 2022-10-14 Mingrui Liu , Zhenxun Zhuang , Yunwei Lei , Chunyang Liao

Compressed communication, in the form of sparsification or quantization of stochastic gradients, is employed to reduce communication costs in distributed data-parallel training of deep neural networks. However, there exists a discrepancy…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-11-20 Aritra Dutta , El Houcine Bergou , Ahmed M. Abdelmoniem , Chen-Yu Ho , Atal Narayan Sahu , Marco Canini , Panos Kalnis

Network consensus optimization has received increasing attention in recent years and has found important applications in many scientific and engineering fields. To solve network consensus optimization problems, one of the most well-known…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-09-10 Xin Zhang , Jia Liu , Zhengyuan Zhu , Elizabeth S. Bentley

Large-scale distributed training is increasingly becoming communication bound. Many gradient compression algorithms have been proposed to reduce the communication overhead and improve scalability. However, it has been observed that in some…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-03-30 Zhuang Wang , Xinyu Wu , T. S. Eugene Ng

Gradient compression is of growing interests for solving constrained optimization problems including compressed sensing, noisy recovery and matrix completion under limited communication resources and storage costs. Convergence analysis of…

Optimization and Control · Mathematics 2024-10-30 Zhaoyue Xia , Jun Du , Chunxiao Jiang , H. Vincent Poor , Yong Ren

Distributed computing offers a high degree of flexibility to accommodate modern learning constraints and the ever increasing size of datasets involved in massive data issues. Drawing inspiration from the theory of distributed computation…

Statistics Theory · Mathematics 2014-07-17 Gérard Biau , Ryad Zenine

In distributed optimization, a large number of machines alternate between local computations and communication with a coordinating server. Communication, which can be slow and costly, is the main bottleneck in this setting. To reduce this…

Machine Learning · Computer Science 2026-04-03 Laurent Condat , Ivan Agarský , Peter Richtárik

We consider machine learning applications that train a model by leveraging data distributed over a trusted network, where communication constraints can create a performance bottleneck. A number of recent approaches propose to overcome this…

Machine Learning · Computer Science 2021-09-10 Osama A. Hanna , Yahya H. Ezzeldin , Christina Fragouli , Suhas Diggavi