paper-with-me

Papers

Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification, and Local Computations

2019-06-06 · Debraj Basu, Deepesh Data, Can Karakus, Suhas Diggavi

Communication bottleneck has been identified as a significant issue in distributed optimization of large-scale learning models. Recently, several approaches to mitigate this problem have been proposed, including different forms of gradient compression or computing local models and mixing them iteratively. In this paper, we propose \emph{Qsparse-local-SGD} algorithm, which combines aggressive sparsification with quantization and local computation along with error compensation, by keeping track of the difference between the true and compressed gradients. We propose both synchronous and asynchronous implementations of \emph{Qsparse-local-SGD}. We analyze convergence for \emph{Qsparse-local-SGD} in the \emph{distributed} setting for smooth non-convex and convex objective functions. We demonstrate that \emph{Qsparse-local-SGD} converges at the same rate as vanilla distributed SGD for many important classes of sparsifiers and quantizers. We use \emph{Qsparse-local-SGD} to train ResNet-50 on ImageNet and show that it results in significant savings over the state-of-the-art, in the number of bits transmitted to reach target accuracy.

📄 PDF Abstract BibTeX arXiv:1906.02367

Code (0)

등록된 구현이 없습니다.

Tasks

Distributed OptimizationQuantization

Methods 이 논문이 사용한 방법론

SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification and Local Computations

2019-12-01 · NeurIPS 2019 12 · Debraj Basu, Deepesh Data, Can Karakus, Suhas Diggavi

Communication bottleneck has been identified as a significant issue in distributed optimization of large-scale learning models. Recently, several approaches to mitigate this problem have been proposed, including differen…

Distributed OptimizationQuantization

Theoretically Better and Numerically Faster Distributed Optimization with Smoothness-Aware Quantization Techniques

2021-06-07 · Bokun Wang, Mher Safaryan, Peter Richtárik

To address the high communication costs of distributed machine learning, a large body of work has been devoted in recent years to designing various compression strategies, such as sparsification and quantization, and opt…

BIG-bench Machine LearningDistributed OptimizationQuantization

JointSQ: Joint Sparsification-Quantization for Distributed Learning

2024-01-01 · CVPR 2024 1 · Weiying Xie, Haowei Li, Jitao Ma, Yunsong Li 외

Gradient sparsification and quantization offer a promising prospect to alleviate the communication overhead problem in distributed learning. However direct combination of the two results in suboptimal solutions due t…

Quantization

LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression

2024-03-07 · Laurent Condat, Artavazd Maranjyan, Peter Richtárik

In Distributed optimization and Learning, and even more in the modern framework of federated learning, communication, which is slow and costly, is critical. We introduce LoCoDL, a communication-efficient algorithm that l…

Distributed OptimizationFederated LearningQuantization

The Convergence of Sparsified Gradient Methods

2018-09-27 · NeurIPS 2018 12 · Dan Alistarh, Torsten Hoefler, Mikael Johansson, Sarit Khirirat 외

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 qua…

Quantization