Skip to main navigation Skip to search Skip to main content

ImageNet training in minutes

  • Yang You
  • , Zhao Zhang
  • , Cho Jui Hsieh
  • , James Demmel
  • , Kurt Keutzer

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

Abstract

In this paper, we investigate large scale computers’ capability of speeding up deep neural networks (DNN) training. Our approach is to use large batch size, powered by the Layer-wise Adaptive Rate Scaling (LARS) algorithm, for efficient usage of massive computing resources. Our approach is generic, as we empirically evaluate the effectiveness on two neural networks: AlexNet and ResNet-50 trained with the ImageNet-1k dataset while preserving the state-of-the-art test accuracy. Compared to the baseline of a previous study from a group of researchers at Facebook, our approach shows higher test accuracy on batch sizes that are larger than 16K. Using 2,048 Intel Xeon Platinum 8160 processors, we reduce the 100-epoch AlexNet training time from hours to 11 minutes. With 2,048 Intel Xeon Phi 7250 Processors, we reduce the 90-epoch ResNet-50 training time from hours to 20 minutes. Our implementation is open source and has been released in the Intel distribution of Caffe v1.0.7.

Original languageEnglish (US)
Title of host publicationProceedings of the 47th International Conference on Parallel Processing, ICPP 2018
PublisherAssociation for Computing Machinery
ISBN (Print)9781450365109
DOIs
StatePublished - Aug 13 2018
Externally publishedYes
Event47th International Conference on Parallel Processing, ICPP 2018 - Eugene, United States
Duration: Aug 14 2018Aug 16 2018

Publication series

NameACM International Conference Proceeding Series

Other

Other47th International Conference on Parallel Processing, ICPP 2018
Country/TerritoryUnited States
CityEugene
Period8/14/188/16/18

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Keywords

  • Distributed machine learning
  • Fast deep neural networks training

Fingerprint

Dive into the research topics of 'ImageNet training in minutes'. Together they form a unique fingerprint.

Cite this