Examining the Effect of Implementation Factors on Deep Learning Reproducibility

Kevin Coakley, Christine R. Kirkpatrick, Odd Erik Gundersen

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

Abstract

Reproducing published deep learning papers to validate their conclusions can be difficult due to sources of irreproducibility. We investigate the impact that implementation factors have on the results and how they affect reproducibility of deep learning studies. Three deep learning experiments were ran five times each on 13 different hardware environments and four different software environments. The analysis of the 780 combined results showed that there was a greater than 6% accuracy range on the same deterministic examples introduced from hardware or software environment variations alone. To account for these implementation factors, researchers should run their experiments multiple times in different hardware and software environments to verify their conclusions are not affected.

Original languageEnglish (US)
Title of host publicationProceedings - 2022 IEEE 18th International Conference on e-Science, eScience 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages397-398
Number of pages2
ISBN (Electronic)9781665461245
DOIs
StatePublished - 2022
Externally publishedYes
Event18th IEEE International Conference on e-Science, eScience 2022 - Salt Lake City, United States
Duration: Oct 10 2022Oct 14 2022

Publication series

NameProceedings - 2022 IEEE 18th International Conference on e-Science, eScience 2022

Conference

Conference18th IEEE International Conference on e-Science, eScience 2022
Country/TerritoryUnited States
CitySalt Lake City
Period10/10/2210/14/22

Keywords

  • deep learning
  • machine learning
  • reproducibility

ASJC Scopus subject areas

  • Library and Information Sciences
  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Signal Processing
  • Modeling and Simulation
  • Instrumentation

Fingerprint

Dive into the research topics of 'Examining the Effect of Implementation Factors on Deep Learning Reproducibility'. Together they form a unique fingerprint.

Cite this