Winning solution on LPIRC-LL competition

Alexander Goncharenko, Sergey Alyamkin, Andrey Denisov, Evgeny Terentev

Результат исследования: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование

Аннотация

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.

Язык оригиналаанглийский
Название основной публикацииProceedings - 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2019
ИздательIEEE Computer Society
Страницы10-16
Число страниц7
ISBN (электронное издание)9781728125060
СостояниеОпубликовано - июн 2019
Событие32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2019 - Long Beach, Соединенные Штаты Америки
Продолжительность: 16 июн 201920 июн 2019

Серия публикаций

НазваниеIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Том2019-June
ISSN (печатное издание)2160-7508
ISSN (электронное издание)2160-7516

Конференция

Конференция32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2019
СтранаСоединенные Штаты Америки
ГородLong Beach
Период16.06.201920.06.2019

Предметные области OECD FOS+WOS

  • 1.02 КОМПЬЮТЕРНЫЕ И ИНФОРМАЦИОННЫЕ НАУКИ
  • 2.02.IQ ИНЖЕНЕРИЯ, ЭЛЕКТРИЧЕСКАЯ И ЭЛЕКТРОННАЯ

Fingerprint

Подробные сведения о темах исследования «Winning solution on LPIRC-LL competition». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать