Adjacent Leader Decentralized Stochastic Gradient Descent

Haoze He, Jing Wang, Anna Choromanska

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This work focuses on the decentralized deep learning optimization framework. We propose Adjacent Leader Decentralized Gradient Descent (AL-DSGD), for improving final model performance, accelerating convergence, and reducing the communication overhead of decentralized deep learning optimizers. AL-DSGD relies on two main ideas. Firstly, to increase the influence of the strongest learners on the learning system it assigns weights to different neighbor workers according to both their performance and the degree when averaging among them, and it applies a corrective force on the workers dictated by both the currently best-performing neighbor and the neighbor with the maximal degree. Secondly, to alleviate the problem of the deterioration of the convergence speed and performance of the nodes with lower degrees, AL-DSGD relies on dynamic communication graphs, which effectively allows the workers to communicate with more nodes while keeping the degrees of the nodes low. Experiments demonstrate that AL-DSGD accelerates the convergence of the decentralized state-of-the-art techniques and improves their test performance especially in the communication constrained environments. We also theoretically prove the convergence of the proposed scheme. Finally, we release to the community a highly general and concise PyTorch-based library for distributed training of deep learning models that supports easy implementation of any distributed deep learning approach ((a)synchronous, (de)centralized).

Original languageEnglish (US)
Title of host publicationECAI 2024 - 27th European Conference on Artificial Intelligence, Including 13th Conference on Prestigious Applications of Intelligent Systems, PAIS 2024, Proceedings
EditorsUlle Endriss, Francisco S. Melo, Kerstin Bach, Alberto Bugarin-Diz, Jose M. Alonso-Moral, Senen Barro, Fredrik Heintz
PublisherIOS Press BV
Pages2492-2499
Number of pages8
ISBN (Electronic)9781643685489
DOIs
StatePublished - Oct 16 2024
Event27th European Conference on Artificial Intelligence, ECAI 2024 - Santiago de Compostela, Spain
Duration: Oct 19 2024Oct 24 2024

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume392
ISSN (Print)0922-6389
ISSN (Electronic)1879-8314

Conference

Conference27th European Conference on Artificial Intelligence, ECAI 2024
Country/TerritorySpain
CitySantiago de Compostela
Period10/19/2410/24/24

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Adjacent Leader Decentralized Stochastic Gradient Descent'. Together they form a unique fingerprint.

Cite this