paper-with-me

홈 › Papers

Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data

2021-02-09 · Tao Lin, Sai Praneeth Karimireddy, Sebastian U. Stich, Martin Jaggi

Decentralized training of deep learning models is a key element for enabling data privacy and on-device learning over networks. In realistic learning scenarios, the presence of heterogeneity across different clients' local datasets poses an optimization challenge and may severely deteriorate the generalization performance. In this paper, we investigate and identify the limitation of several decentralized optimization algorithms for different degrees of data heterogeneity. We propose a novel momentum-based method to mitigate this decentralized training difficulty. We show in extensive empirical experiments on various CV/NLP datasets (CIFAR-10, ImageNet, and AG News) and several network topologies (Ring and Social Network) that our method is much more robust to the heterogeneity of clients' data than other existing methods, by a significant improvement in test performance ($1\% \!-\! 20\%$). Our code is publicly available.

📄 PDF Abstract BibTeX arXiv:2102.04761

Code (1)

epfml/quasi-global-momentum 공식 구현 pytorch

Tasks

Deep Learning

Similar Papers 제목 키워드 기반

Communication-Efficient Neural Tangent Kernels for Heterogeneous Decentralized Federated Learning

2025-12-14 · Li Xia arxiv

Decentralized federated learning (DFL) enables collaborative model training without a central server, but converges slowly under statistical heterogeneity. Recent work has shown that neural tangent kernel (NTK) methods a…

Federated Learning

Momentum Tracking: Momentum Acceleration for Decentralized Deep Learning on Heterogeneous Data

2022-09-30 · Yuki Takezawa, Han Bao, Kenta Niwa, Ryoma Sato 외

SGD with momentum is one of the key components for improving the performance of neural networks. For decentralized learning, a straightforward approach using momentum is Distributed SGD (DSGD) with momentum (DSGDm). Howe…

Image Classification

A Unified Momentum-based Paradigm of Decentralized SGD for Non-Convex Models and Heterogeneous Data

2023-03-01 · Haizhou Du, Chengdong Ni

Emerging distributed applications recently boosted the development of decentralized machine learning, especially in IoT and edge computing fields. In real-world scenarios, the common problems of non-convexity and data he…

Edge-computingglobal-optimization

Ordered Local Momentum for Asynchronous Distributed Learning under Arbitrary Delays

2026-01-18 · Chang-Wei Shi, Shi-Shang Wang, Wu-Jun Li arxiv

Momentum SGD (MSGD) serves as a foundational optimizer in training deep models due to momentum's key role in accelerating convergence and enhancing generalization. Meanwhile, asynchronous distributed learning is crucial …

DecentLaM: Decentralized Momentum SGD for Large-batch Deep Training

2021-04-24 · ICCV 2021 10 · Kun Yuan, Yiming Chen, Xinmeng Huang, Yingya Zhang 외

The scale of deep learning nowadays calls for efficient distributed training algorithms. Decentralized momentum SGD (DmSGD), in which each node averages only with its neighbors, is more communication efficient than vanil…