VaB-AL: Incorporating Class Imbalance and Difficulty with Variational Bayes for Active Learning
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Choi, Jongwon | - |
dc.contributor.author | Yi, K.M. | - |
dc.contributor.author | Kim, J. | - |
dc.contributor.author | Choo, J. | - |
dc.contributor.author | Kim, B. | - |
dc.contributor.author | Chang, J. | - |
dc.contributor.author | Gwon, Y. | - |
dc.contributor.author | Chang, H.J. | - |
dc.date.accessioned | 2023-03-08T10:06:23Z | - |
dc.date.available | 2023-03-08T10:06:23Z | - |
dc.date.issued | 2021-11 | - |
dc.identifier.issn | 1063-6919 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/62063 | - |
dc.description.abstract | Active Learning for discriminative models has largely been studied with the focus on individual samples, with less emphasis on how classes are distributed or which classes are hard to deal with. In this work, we show that this is harmful. We propose a method based on the Bayes' rule, that can naturally incorporate class imbalance into the Active Learning framework. We derive that three terms should be considered together when estimating the probability of a classifier making a mistake for a given sample; i) probability of mislabelling a class, ii) likelihood of the data given a predicted class, and iii) the prior probability on the abundance of a predicted class. Implementing these terms requires a generative model and an intractable likelihood estimation. Therefore, we train a Variational Auto Encoder (VAE) for this purpose. To further tie the VAE with the classifier and facilitate VAE training, we use the classifiers' deep feature representations as input to the VAE. By considering all three probabilities, among them, especially the data imbalance, we can substantially improve the potential of existing methods under limited data budget. We show that our method can be applied to classification tasks on multiple different datasets - including one that is a real-world dataset with heavy data imbalance - significantly outperforming the state of the art. © 2021 IEEE | - |
dc.format.extent | 10 | - |
dc.language | 영어 | - |
dc.language.iso | ENG | - |
dc.publisher | IEEE Computer Society | - |
dc.title | VaB-AL: Incorporating Class Imbalance and Difficulty with Variational Bayes for Active Learning | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/CVPR46437.2021.00668 | - |
dc.identifier.bibliographicCitation | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp 6745 - 6754 | - |
dc.description.isOpenAccess | N | - |
dc.identifier.wosid | 000739917306095 | - |
dc.identifier.scopusid | 2-s2.0-85123213853 | - |
dc.citation.endPage | 6754 | - |
dc.citation.startPage | 6745 | - |
dc.citation.title | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | - |
dc.type.docType | Proceedings Paper | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Imaging Science & Photographic Technology | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Artificial Intelligence | - |
dc.relation.journalWebOfScienceCategory | Imaging Science & Photographic Technology | - |
dc.description.journalRegisteredClass | scopus | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
84, Heukseok-ro, Dongjak-gu, Seoul, Republic of Korea (06974)02-820-6194
COPYRIGHT 2019 Chung-Ang University All Rights Reserved.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.