Asynchronous Decentralized Bayesian Optimization for Large Scale Hyperparameter Optimization

Romain Egelé, Isabelle Guyon, Venkatram Vishwanath, Prasanna Balaprakash

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Bayesian optimization (BO) is a promising approach for hyperparameter optimization of deep neural networks (DNNs), where each model training can take minutes to hours. In BO, a computationally cheap surrogate model is employed to learn the relationship between parameter configurations and their performance such as accuracy. Parallel BO methods often adopt single manager/multiple workers strategies to evaluate multiple hyperparameter configurations simultaneously. Despite significant hyperparameter evaluation time, the overhead in such centralized schemes prevents these methods to scale on a large number of workers. We present an asynchronous-decentralized BO, wherein each worker runs a sequential BO and asynchronously communicates its results through shared storage. We scale our method without loss of computational efficiency with above 95% of worker's utilization to 1,920 parallel workers (full production queue of the Polaris supercomputer) and demonstrate improvement in model accuracy as well as faster convergence on the CANDLE benchmark from the Exascale computing project.

Original languageEnglish
Title of host publicationProceedings 2023 IEEE 19th International Conference on e-Science, e-Science 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350322231
DOIs
StatePublished - 2023
Event19th IEEE International Conference on e-Science, e-Science 2023 - Limassol, Cyprus
Duration: Oct 9 2023Oct 14 2023

Publication series

NameProceedings 2023 IEEE 19th International Conference on e-Science, e-Science 2023

Conference

Conference19th IEEE International Conference on e-Science, e-Science 2023
Country/TerritoryCyprus
CityLimassol
Period10/9/2310/14/23

Keywords

  • Bayesian optimization
  • asynchronous parallel computing
  • hyperparameter optimization
  • machine learning

Fingerprint

Dive into the research topics of 'Asynchronous Decentralized Bayesian Optimization for Large Scale Hyperparameter Optimization'. Together they form a unique fingerprint.

Cite this