Removing Data Heterogeneity Influence Enhances Network Topology Dependence of Decentralized SGD

17 May 2021

Papers citing "Removing Data Heterogeneity Influence Enhances Network Topology Dependence of Decentralized SGD"

9 / 9 papers shown

Title
Convergence of Decentralized Stochastic Subgradient-based Methods for Nonsmooth Nonconvex functions Siyuan Zhang Nachuan Xiao Xin Liu 61 1 0 18 Mar 2024
Distributed Random Reshuffling Methods with Improved Convergence Kun-Yen Huang Linli Zhou Shi Pu 24 4 0 21 Jun 2023
CEDAS: A Compressed Decentralized Stochastic Gradient Method with Improved Convergence Kun-Yen Huang Shin-Yi Pu 30 9 0 14 Jan 2023
Data-heterogeneity-aware Mixing for Decentralized Learning Yatin Dandi Anastasia Koloskova Martin Jaggi Sebastian U. Stich 38 18 0 13 Apr 2022
BlueFog: Make Decentralized Algorithms Practical for Optimization and Deep Learning Bicheng Ying Kun Yuan Hanbin Hu Yiming Chen W. Yin FedML 31 27 0 08 Nov 2021
Exponential Graph is Provably Efficient for Decentralized Deep Training Bicheng Ying Kun Yuan Yiming Chen Hanbin Hu Pan Pan W. Yin FedML 39 83 0 26 Oct 2021
A Unified and Refined Convergence Analysis for Non-Convex Decentralized Learning Sulaiman A. Alghunaim Kun Yuan 25 57 0 19 Oct 2021
DecentLaM: Decentralized Momentum SGD for Large-batch Deep Training Kun Yuan Yiming Chen Xinmeng Huang Yingya Zhang Pan Pan Yinghui Xu W. Yin MoE 55 61 0 24 Apr 2021
Distributed Pareto Optimization via Diffusion Strategies Jianshu Chen Ali H. Sayed 71 174 0 13 Aug 2012