# Straggler-Resilient Differentially-Private Decentralized Learning

Source: [Straggler-Resilient Differentially-Private Decentralized Learning](<https://www.sota2.com/research/paper/68fa5c011e209131ce9cd664>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:56:09.998Z

Authors: Yauhen Yakimenka, Chung-Wei Weng, Hsuan-Yin Lin, Eirik Rosnes, J\\"org Kliewer

Published: 2022-12-06T15:50:28.000Z

## Abstract

We consider the straggler problem in decentralized learning over a logical ring while preserving user data privacy. Especially, we extend the recently proposed framework of differential privacy (DP) amplification by decentralization by Cyffers and Bellet to include overall training latency--comprising both computation and communication latency. Analytical results on both the convergence speed and the DP level are derived for both a skipping scheme (which ignores the stragglers after a timeout) and a baseline scheme that waits for each node to finish before the training continues. A trade-off between overall training latency, accuracy, and privacy, parameterized by the timeout of the skipping scheme, is identified and empirically validated for logistic regression on a real-world dataset and for image classification using the MNIST and CIFAR-10 datasets.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2212.03080>)

[PDF](<https://arxiv.org/pdf/2212.03080.pdf>)