Dual-Classifier Collaborative Method Based on Semi-Supervised Active Learning

Baoguo Wei, Xinyu Wang, Yue Zhang, Mingzhi Cai, Xu Li, Lixin Li

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

Abstract

Active Learning aims to reduce the annotation cost by selecting the most informative and representative samples to label. Typical active learning methods tend to overlook information from unlabeled samples. In contrast, semi-supervised learning utilizes unlabeled data that the model considers to be correctly classified temporarily. By integrating the advantages of both methods, we propose dual-classifier collaborative method based on semi-supervised active learning. Specifically, samples with the largest prediction difference between dual classifiers are queried for labeling by degree of prediction difference, while those with high confidence receive pseudo labels informed by the two classifiers. Under the increasing of labeled data, experiments on CIFAR and SVNH datasets show our method outperforms existing methods by 0.8% 5.84% in terms of classification accuracy.

Original languageEnglish
Title of host publicationProceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350316728
DOIs
StatePublished - 2023
Event2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023 - Zhengzhou, Henan, China
Duration: 14 Nov 202317 Nov 2023

Publication series

NameProceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

Conference

Conference2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023
Country/TerritoryChina
CityZhengzhou, Henan
Period14/11/2317/11/23

Keywords

  • active learning
  • dual classifiers
  • semi-supervised learning

Fingerprint

Dive into the research topics of 'Dual-Classifier Collaborative Method Based on Semi-Supervised Active Learning'. Together they form a unique fingerprint.

Cite this