Gaussian Based Active Learning Algorithm for Image Classification Problem

Andrey Shcherbin, Gulnara Yakhyaeva

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

Abstract

One of the relevant problems in deep learning is data efficiency. In the active learning approach, we have a large set of unlabeled data, a small set of labeled data and a limited budget for labeling. The model for training on labeled data is defined. The task is to select the most relevant samples to increase model quality on the test data set. In this work, we overview some active learning algorithms and propose a novel algorithm for active learning based on Gaussian distribution. The main idea of the algorithm is to use reference samples from each class and compute the distribution parameters (and) for each embedding coordinate. We use Gaussian function as a measure of distance between unlabeled samples and class representation, so we can combine it with any confidence-based algorithm. We tested our approach on a part of ImageNet task (20 random classes from original ImageNet 2012 dataset). We used a Gaussian distribution-based measure combined with a confidence-based sample selection method. This combination achieved significantly better results.

Original languageEnglish
Title of host publication2021 IEEE 22nd International Conference of Young Professionals in Electron Devices and Materials, EDM 2021 - Proceedings
PublisherIEEE Computer Society
Pages542-546
Number of pages5
ISBN (Electronic)9781665414982
DOIs
Publication statusPublished - 30 Jun 2021
Event22nd IEEE International Conference of Young Professionals in Electron Devices and Materials, EDM 2021 - Aya, Altai Region, Russian Federation
Duration: 30 Jun 20214 Jul 2021

Publication series

NameInternational Conference of Young Specialists on Micro/Nanotechnologies and Electron Devices, EDM
Volume2021-June
ISSN (Print)2325-4173
ISSN (Electronic)2325-419X

Conference

Conference22nd IEEE International Conference of Young Professionals in Electron Devices and Materials, EDM 2021
CountryRussian Federation
CityAya, Altai Region
Period30.06.202104.07.2021

Keywords

  • active learning
  • Gaussian filtering
  • least confidence
  • machine learning
  • margin sampling
  • maximum entropy

OECD FOS+WOS

  • 2.02.IQ ENGINEERING, ELECTRICAL & ELECTRONIC
  • 1.03.UH PHYSICS, ATOMIC, MOLECULAR & CHEMICAL
  • 1.03.SY OPTICS

Fingerprint

Dive into the research topics of 'Gaussian Based Active Learning Algorithm for Image Classification Problem'. Together they form a unique fingerprint.

Cite this