English
Related papers

Related papers: Quantize Once, Train Fast: Allreduce-Compatible Co…

200 papers

Gradient compression has surfaced as a key technique to address the challenge of communication efficiency in distributed learning. In distributed deep learning, however, it is observed that gradient distributions are heavy-tailed, with…

Machine Learning · Computer Science 2024-02-07 Guangfeng Yan , Tan Li , Yuanzhang Xiao , Hanxu Hou , Linqi Song

Federated learning is a promising framework to mitigate data privacy and computation concerns. However, the communication cost between the server and clients has become the major bottleneck for successful deployment. Despite notable…

Machine Learning · Computer Science 2022-05-03 Yang He , Hui-Po Wang , Maximilian Zenk , Mario Fritz

We study gradient compression methods to alleviate the communication bottleneck in data-parallel distributed optimization. Despite the significant attention received, current compression schemes either do not scale well or fail to achieve…

Machine Learning · Computer Science 2020-02-19 Thijs Vogels , Sai Praneeth Karimireddy , Martin Jaggi

Distributed training is an effective way to accelerate the training process of large-scale deep learning models. However, the parameter exchange and synchronization of distributed stochastic gradient descent introduce a large amount of…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-08-16 LingFei Dai , Boyu Diao , Chao Li , Yongjun Xu

Data explosion and an increase in model size drive the remarkable advances in large-scale machine learning, but also make model training time-consuming and model storage difficult. To address the above issues in the distributed model…

Machine Learning · Computer Science 2022-08-12 Ke Xu , Jianqiao Wangni , Yifan Zhang , Deheng Ye , Jiaxiang Wu , Peilin Zhao

Quantized deep neural networks (QDNNs) are attractive due to their much lower memory storage and faster inference speed than their regular full precision counterparts. To maintain the same performance level especially at low bit-widths,…

Machine Learning · Computer Science 2019-01-08 Penghang Yin , Shuai Zhang , Jiancheng Lyu , Stanley Osher , Yingyong Qi , Jack Xin

Communication is a crucial phase in the context of distributed training. Because parameter server (PS) frequently experiences network congestion, recent studies have found that training paradigms without a centralized server outperform the…

Optimization and Control · Mathematics 2020-12-17 Feijie Wu , Shiqi He , Yutong Yang , Haozhao Wang , Zhihao Qu , Song Guo , Weihua Zhuang

Multi-hop all-reduce is the de facto backbone of large model training. As the training scale increases, the network often becomes a bottleneck, motivating reducing the volume of transmitted data. Accordingly, recent systems demonstrated…

Machine Learning · Computer Science 2026-02-10 Wenchen Han , Shay Vargaftik , Michael Mitzenmacher , Ran Ben Basat

Lossy gradient compression, with either unbiased or biased compressors, has become a key tool to avoid the communication bottleneck in centrally coordinated distributed training of machine learning models. We analyze the performance of two…

Machine Learning · Computer Science 2020-12-23 Sebastian U. Stich

Large-scale deep neural networks (DNN) exhibit excellent performance for various tasks. As DNNs and datasets grow, distributed training becomes extremely time-consuming and demands larger clusters. A main bottleneck is the resulting…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-05-27 Yisu Wang , Ruilong Wu , Xinjiao Li , Dirk Kutscher

One of the most significant bottleneck in training large scale machine learning models on parameter server (PS) is the communication overhead, because it needs to frequently exchange the model gradients between the workers and servers…

Machine Learning · Computer Science 2018-04-25 Guoxin Cui , Jun Xu , Wei Zeng , Yanyan Lan , Jiafeng Guo , Xueqi Cheng

In distributed training, communication often emerges as a bottleneck. In response, we introduce Kimad, a solution that offers adaptive gradient compression. By consistently monitoring bandwidth, Kimad refines compression ratios to match…

Machine Learning · Computer Science 2023-12-14 Jihao Xin , Ivan Ilin , Shunkang Zhang , Marco Canini , Peter Richtárik

Training generative adversarial networks (GAN) in a distributed fashion is a promising technology since it is contributed to training GAN on a massive of data efficiently in real-world applications. However, GAN is known to be difficult to…

Machine Learning · Computer Science 2020-10-27 Xiaojun Chen , Shu Yang , Li Shen , Xuanrong Pang

Gradient aggregation has long been identified as a major bottleneck in today's large-scale distributed machine learning training systems. One promising solution to mitigate such bottlenecks is gradient compression, directly reducing…

Machine Learning · Computer Science 2024-10-30 Wenchen Han , Shay Vargaftik , Michael Mitzenmacher , Brad Karp , Ran Ben Basat

Low-rank gradient compression methods, such as PowerSGD, have gained attention in communication-efficient distributed optimization. However, the convergence guarantees of PowerSGD remain unclear, particularly in stochastic settings. In this…

Optimization and Control · Mathematics 2025-09-16 Shengping Xie , Chuyan Chen , Kun Yuan

Communication compression is a crucial technique for modern distributed learning systems to alleviate their communication bottlenecks over slower networks. Despite recent intensive studies of gradient compression for data parallel-style…

Machine Learning · Computer Science 2023-03-08 Jue Wang , Binhang Yuan , Luka Rimanic , Yongjun He , Tri Dao , Beidi Chen , Christopher Re , Ce Zhang

To accelerate distributed training, many gradient compression methods have been proposed to alleviate the communication bottleneck in synchronous stochastic gradient descent (S-SGD), but their efficacy in real-world applications still…

Machine Learning · Computer Science 2023-06-16 Lin Zhang , Longteng Zhang , Shaohuai Shi , Xiaowen Chu , Bo Li

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…

We consider a decentralized learning problem, where a set of computing nodes aim at solving a non-convex optimization problem collaboratively. It is well-known that decentralized optimization schemes face two major system bottlenecks:…

Machine Learning · Computer Science 2019-11-04 Amirhossein Reisizadeh , Hossein Taheri , Aryan Mokhtari , Hamed Hassani , Ramtin Pedarsani

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