Winning solution on LPIRC-LL competition

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

Abstract

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain the accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNetv2 and MNAS. Here we present two methods to significantly optimize the training with quantization procedure. The first one is introducing the trained scale factors for discretization thresholds that are separate for each filter. The second one is based on mutual rescaling of consequent depth-wise separable convolution and convolution layers. Using the proposed techniques, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available at: https://github.com/agoncharenko1992/FAT-fastadjustable-threshold.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2019
PublisherIEEE Computer Society
Pages10-16
Number of pages7
ISBN (Electronic)9781728125060
Publication statusPublished - Jun 2019
Event32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2019 - Long Beach, United States
Duration: 16 Jun 201920 Jun 2019

Publication series

NameIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Volume2019-June
ISSN (Print)2160-7508
ISSN (Electronic)2160-7516

Conference

Conference32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2019
CountryUnited States
CityLong Beach
Period16.06.201920.06.2019

OECD FOS+WOS

  • 1.02 COMPUTER AND INFORMATION SCIENCES
  • 2.02.IQ ENGINEERING, ELECTRICAL & ELECTRONIC

Fingerprint

Dive into the research topics of 'Winning solution on LPIRC-LL competition'. Together they form a unique fingerprint.

Cite this