paper-with-me

Papers

CD-SGD: Distributed Stochastic Gradient Descent with Compression and Delay Compensation

2021-06-21 · Enda Yu, Dezun Dong, Yemao Xu, Shuo Ouyang, Xiangke Liao

Communication overhead is the key challenge for distributed training. Gradient compression is a widely used approach to reduce communication traffic. When combining with parallel communication mechanism method like pipeline, gradient compression technique can greatly alleviate the impact of communication overhead. However, there exists two problems of gradient compression technique to be solved. Firstly, gradient compression brings in extra computation cost, which will delay the next training iteration. Secondly, gradient compression usually leads to the decrease of convergence accuracy.

📄 PDF Abstract BibTeX arXiv:2106.10796

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

A Tight Convergence Analysis for Stochastic Gradient Descent with Delayed Updates

2018-06-26 · Yossi Arjevani, Ohad Shamir, Nathan Srebro

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…

Distributed Optimization

Flattened one-bit stochastic gradient descent: compressed distributed optimization with controlled variance

2024-05-17 · Alexander Stollenwerk, Laurent Jacques

We propose a novel algorithm for distributed stochastic gradient descent (SGD) with compressed gradient communication in the parameter-server framework. Our gradient compression technique, named flattened one-bit stochas…

Distributed OptimizationQuantization

Delay-Tolerant Local SGD for Efficient Distributed Training

2021-01-01 · An Xu, Xiao Yan, Hongchang Gao, Heng Huang

The heavy communication for model synchronization is a major bottleneck for scaling up the distributed deep neural network training to many workers. Moreover, model synchronization can suffer from long delays in scenario…

Federated Learning

Distributed stochastic optimization with large delays

2021-07-06 · Zhengyuan Zhou, Panayotis Mertikopoulos, Nicholas Bambos, Peter W. Glynn 외

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 stochast…

Distributed ComputingStochastic Optimization

On Biased Compression for Distributed Learning

2020-02-27 · Aleksandr Beznosikov, Samuel Horváth, Peter Richtárik, Mher Safaryan

In the last few years, various communication compression techniques have emerged as an indispensable tool helping to alleviate the communication bottleneck in distributed learning. However, despite the fact biased compre…