Skip to main navigation Skip to search Skip to main content

Efficient Training of EfficientNetV2-S Using AdaBelief Optimizer

  • Sungkyunkwan University

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

Abstract

The EfficientNetV2 architectures have classification accuracies exceeding 80% on the Imagenet-1K/21K datasets. Here, the networks used to be trained using stochastic gradient descent with momentum ($SGD+M$) as the optimizer. Such a method is known to be effective but takes too much time, making it impractical to train the networks from scratch especially when computational power is limited. To address this, we provide a guide on how to use the AdaBelief Optimizer for training EfficientNetV2-S. Results show that, even without complex training configurations, using AdaBelief can lead the EfficientNetV2-S network to achieve a top-1 accuracy as high as 80% on the Imagenet-1K validation set within 80 epochs of training.

Original languageEnglish
Title of host publication2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665464345
DOIs
StatePublished - 2022
Externally publishedYes
Event2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022 - Yeosu, Korea, Republic of
Duration: 26 Oct 202228 Oct 2022

Publication series

Name2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022

Conference

Conference2022 IEEE International Conference on Consumer Electronics-Asia, ICCE-Asia 2022
Country/TerritoryKorea, Republic of
CityYeosu
Period26/10/2228/10/22

Keywords

  • AdaBelief
  • deep learning
  • EfficientNetV2-S
  • image classification

Fingerprint

Dive into the research topics of 'Efficient Training of EfficientNetV2-S Using AdaBelief Optimizer'. Together they form a unique fingerprint.

Cite this