RSG: A Simple but Effective Module for Learning Imbalanced Datasets

Imbalanced datasets widely exist in practice and are a great challenge for training deep neural models with a good generalization on infrequent classes. In this work, we propose a new rare-class sample generator (RSG) to solve this problem. RSG aims to generate some new samples for rare classes duri...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 3783 - 3792
Hauptverfasser: Wang, Jianfeng, Lukasiewicz, Thomas, Hu, Xiaolin, Cai, Jianfei, Xu, Zhenghua
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2021
Schlagworte:
ISSN:1063-6919
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Imbalanced datasets widely exist in practice and are a great challenge for training deep neural models with a good generalization on infrequent classes. In this work, we propose a new rare-class sample generator (RSG) to solve this problem. RSG aims to generate some new samples for rare classes during training, and it has in particular the following advantages: (1) it is convenient to use and highly versatile, because it can be easily integrated into any kind of convolutional neural network, and it works well when combined with different loss functions, and (2) it is only used during the training phase, and therefore, no additional burden is imposed on deep neural networks during the testing phase. In extensive experimental evaluations, we verify the effectiveness of RSG. Furthermore, by leveraging RSG, we obtain competitive results on Imbalanced CIFAR and new state-of-the-art results on Places-LT, ImageNet-LT, and iNaturalist 2018. The source code is available at https://github.com/Jianf-Wang/RSG.
AbstractList Imbalanced datasets widely exist in practice and are a great challenge for training deep neural models with a good generalization on infrequent classes. In this work, we propose a new rare-class sample generator (RSG) to solve this problem. RSG aims to generate some new samples for rare classes during training, and it has in particular the following advantages: (1) it is convenient to use and highly versatile, because it can be easily integrated into any kind of convolutional neural network, and it works well when combined with different loss functions, and (2) it is only used during the training phase, and therefore, no additional burden is imposed on deep neural networks during the testing phase. In extensive experimental evaluations, we verify the effectiveness of RSG. Furthermore, by leveraging RSG, we obtain competitive results on Imbalanced CIFAR and new state-of-the-art results on Places-LT, ImageNet-LT, and iNaturalist 2018. The source code is available at https://github.com/Jianf-Wang/RSG.
Author Xu, Zhenghua
Lukasiewicz, Thomas
Cai, Jianfei
Hu, Xiaolin
Wang, Jianfeng
Author_xml – sequence: 1
  givenname: Jianfeng
  surname: Wang
  fullname: Wang, Jianfeng
  email: jianfeng.wang@cs.ox.ac.uk
  organization: University of Oxford
– sequence: 2
  givenname: Thomas
  surname: Lukasiewicz
  fullname: Lukasiewicz, Thomas
  email: thomas.lukasiewicz@cs.ox.ac.uk
  organization: University of Oxford
– sequence: 3
  givenname: Xiaolin
  surname: Hu
  fullname: Hu, Xiaolin
  email: xlhu@tsinghua.edu.cn
  organization: Tsinghua University
– sequence: 4
  givenname: Jianfei
  surname: Cai
  fullname: Cai, Jianfei
  email: jianfei.cai@monash.edu
  organization: Monash University
– sequence: 5
  givenname: Zhenghua
  surname: Xu
  fullname: Xu, Zhenghua
  email: zhenghua.xu@hebut.edu.cn
  organization: Hebei University of Technology
BookMark eNotjMtKw0AUQEdRsK39Al3MD6TeO8-MuxJrLUSUVt2WO8mNRNK0JKng31vQ1YHD4YzFRbtvWYhbhBkihLvs43VtnNF-pkDhDED79EyM0TlrjIWgzsUIwenEBQxXYtr3X3CKFKIL6Uhk683yXs7lpt4dGpbxOMhFVXEx1N8sn_fl8SSrfSdzpq6t20-52kVqqC24lA80UM9Dfy0uK2p6nv5zIt4fF2_ZU5K_LFfZPE9qBXpIIkbN5MAWqbdYlYo0OwLtovHOFQq1CkZHhx6LSAYMp5bKoCIQoLesJ-Lm71sz8_bQ1TvqfrbB-lQbpX8BV3NLFg
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR46437.2021.00378
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
EISBN 1665445092
9781665445092
EISSN 1063-6919
EndPage 3792
ExternalDocumentID 9578342
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i203t-b1b3ea605c8751fd2a3e6a036b4766c2132943b6171cba404e85ad92b0a0175e3
IEDL.DBID RIE
ISICitedReferencesCount 78
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000739917303096&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:28:30 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-b1b3ea605c8751fd2a3e6a036b4766c2132943b6171cba404e85ad92b0a0175e3
PageCount 10
ParticipantIDs ieee_primary_9578342
PublicationCentury 2000
PublicationDate 2021-June
PublicationDateYYYYMMDD 2021-06-01
PublicationDate_xml – month: 06
  year: 2021
  text: 2021-June
PublicationDecade 2020
PublicationTitle Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online)
PublicationTitleAbbrev CVPR
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0003211698
Score 2.5418792
Snippet Imbalanced datasets widely exist in practice and are a great challenge for training deep neural models with a good generalization on infrequent classes. In...
SourceID ieee
SourceType Publisher
StartPage 3783
SubjectTerms Codes
Computer vision
Convolutional neural networks
Deep learning
Generators
Pattern recognition
Training
Title RSG: A Simple but Effective Module for Learning Imbalanced Datasets
URI https://ieeexplore.ieee.org/document/9578342
WOSCitedRecordID wos000739917303096&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwELVKxcBUoEV8ywMjaeM4iW02VCggQVW1gLpV_rigDm1Rm_D7OSdRERILm-XhLJ1lvzv73jtCrrTNEmdEFiTKhYHXwgy0cipgAhLJIAMJJVH4WQyHcjpVowa53nJhAKAsPoOuH5Z_-W5lC_9U1lOJbwuBF-6OEGnF1dq-p3DMZFIla3YcC1Wv_z4ax_5fCrPAiHW90or81UOlhJBB63-L75PODxePjrYoc0AasDwkrTp4pPXR3LRJfzx5uKG3dDL3er_UFDmtlInxOqMvK1fgJAaotBZU_aBPC-PLGi1audM5glm-6ZC3wf1r_zGoOyQE8yjkeWCY4aAxI7GYdrDMRZpDqhGUTCzS1Ea-i3zMDUYpzBodhzHIRDsVmVDjSUyAH5HmcrWEY0JlqFyM9tBWGmchl1Y5oQDxXGmTcHNC2t4ns89KBGNWu-P07-kzsuedXtVUnZNmvi7gguzar3y-WV-WO_cN_5GYoQ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEG2ImugJFYzf9uDRhe623W29GRQhAiGAhhvpx6zhIBjY9ffbLhuMiRdvTQ_TpE3nzbTz3iB0q0zKrU7SgEtLAq-FGShpZRAmwEUIKQgoiMK9ZDAQ06kcVtDdlgsDAEXxGTT8sPjLt0uT-6eypuS-LYRzuLucsYhs2FrbFxXqcplYipIfFxLZbL0NR8z_TLk8MAobXmtF_OqiUoBIu_q_5Q9R_YeNh4dbnDlCFVgco2oZPuLycq5rqDUaP9_jBzyee8VfrPMMb7SJnUPD_aXN3aQLUXEpqfqOux_aFzYaZ-VRZQ7OsnUdvbafJq1OUPZICOYRoVmgQ01BuZzEuMQjTG2kKMTKwZJmSRybyPeRZ1S7OCU0WjHCQHBlZaSJcneRAz1BO4vlAk4RFkRa5uw5WzFLCRVG2kSCQ3SpNKf6DNX8nsw-NzIYs3I7zv-evkH7nUm_N-t1By8X6MAfwKbC6hLtZKscrtCe-crm69V1cYrfI-Cb6A
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=proceeding&rft.title=Proceedings+%28IEEE+Computer+Society+Conference+on+Computer+Vision+and+Pattern+Recognition.+Online%29&rft.atitle=RSG%3A+A+Simple+but+Effective+Module+for+Learning+Imbalanced+Datasets&rft.au=Wang%2C+Jianfeng&rft.au=Lukasiewicz%2C+Thomas&rft.au=Hu%2C+Xiaolin&rft.au=Cai%2C+Jianfei&rft.date=2021-06-01&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=3783&rft.epage=3792&rft_id=info:doi/10.1109%2FCVPR46437.2021.00378&rft.externalDocID=9578342