Averaging Rate Scheduler for Decentralized Learning on Heterogeneous Data

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Aketi, Sai Aparna, Choudhary, Sakshi, Roy, Kaushik
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909129382035456
author Aketi, Sai Aparna
Choudhary, Sakshi
Roy, Kaushik
author_facet Aketi, Sai Aparna
Choudhary, Sakshi
Roy, Kaushik
contents State-of-the-art decentralized learning algorithms typically require the data distribution to be Independent and Identically Distributed (IID). However, in practical scenarios, the data distribution across the agents can have significant heterogeneity. In this work, we propose averaging rate scheduling as a simple yet effective way to reduce the impact of heterogeneity in decentralized learning. Our experiments illustrate the superiority of the proposed method (~3% improvement in test accuracy) compared to the conventional approach of employing a constant averaging rate.
format Preprint
id arxiv_https___arxiv_org_abs_2403_03292
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Averaging Rate Scheduler for Decentralized Learning on Heterogeneous Data
Aketi, Sai Aparna
Choudhary, Sakshi
Roy, Kaushik
Machine Learning
Distributed, Parallel, and Cluster Computing
State-of-the-art decentralized learning algorithms typically require the data distribution to be Independent and Identically Distributed (IID). However, in practical scenarios, the data distribution across the agents can have significant heterogeneity. In this work, we propose averaging rate scheduling as a simple yet effective way to reduce the impact of heterogeneity in decentralized learning. Our experiments illustrate the superiority of the proposed method (~3% improvement in test accuracy) compared to the conventional approach of employing a constant averaging rate.
title Averaging Rate Scheduler for Decentralized Learning on Heterogeneous Data
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2403.03292