Skip to main navigation Skip to search Skip to main content

Short Paper: Accelerating Hyperparameter Optimization Algorithms with Mixed Precision

  • Marcel Aach
  • , Rakesh Sarma
  • , Eray Inanc
  • , Morris Riedel
  • , Andreas Lintermann

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

Abstract

Hyperparameter Optimization (HPO) of Neural Networks (NNs) is a computationally expensive procedure. On accelerators, such as NVIDIA Graphics Processing Units (GPUs) equipped with Tensor Cores, it is possible to speed-up the NN training by reducing the precision of some of the NN parameters, also referred to as mixed precision training. This paper investigates the performance of three popular HPO algorithms in terms of the achieved speed-up and model accuracy, utilizing early stopping, Bayesian, and genetic optimization approaches, in combination with mixed precision functionalities. The benchmarks are performed on 64 GPUs in parallel on three datasets: two from the vision and one from the Computational Fluid Dynamics domain. The results show that larger speed-ups can be achieved for mixed compared to full precision HPO if the checkpoint frequency is kept low. In addition to the reduced runtime, small gains in generalization performance on the test set are observed.

Original languageEnglish
Title of host publicationProceedings of 2023 SC Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023
PublisherAssociation for Computing Machinery, Inc
Pages1776-1779
Number of pages4
ISBN (Electronic)9798400707858
DOIs
Publication statusPublished - 12 Nov 2023
Event2023 International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023 - Denver, United States
Duration: 12 Nov 202317 Nov 2023

Publication series

NameProceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis

Conference

Conference2023 International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023
Country/TerritoryUnited States
CityDenver
Period12/11/2317/11/23

Bibliographical note

Publisher Copyright: © 2023 ACM.

Other keywords

  • High-Performance Computing
  • Hyperparameter Optimization
  • Mixed Precision

Fingerprint

Dive into the research topics of 'Short Paper: Accelerating Hyperparameter Optimization Algorithms with Mixed Precision'. Together they form a unique fingerprint.

Cite this