Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder

State of the art (SOTA) few-shot learning (FSL) methods suffer significant performance drop in the presence of domain differences between source and target datasets. The strong discrimination ability on the source dataset does not necessarily translate to high classification accuracy on the target d...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings / IEEE International Conference on Computer Vision pp. 9404 - 9414
Main Authors: Liang, Hanwen, Zhang, Qiong, Dai, Peng, Lu, Juwei
Format: Conference Proceeding
Language:English
Published: IEEE 01.10.2021
Subjects:
ISSN:2380-7504
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract State of the art (SOTA) few-shot learning (FSL) methods suffer significant performance drop in the presence of domain differences between source and target datasets. The strong discrimination ability on the source dataset does not necessarily translate to high classification accuracy on the target dataset. In this work, we address this cross-domain few-shot learning (CDFSL) problem by boosting the generalization capability of the model. Specifically, we teach the model to capture broader variations of the feature distributions with a novel noise-enhanced supervised autoencoder (NSAE). NSAE trains the model by jointly reconstructing inputs and predicting the labels of inputs as well as their reconstructed pairs. Theoretical analysis based on intra-class correlation (ICC) shows that the feature embeddings learned from NSAE have stronger discrimination and generalization abilities in the target domain. We also take advantage of NSAE structure and propose a two-step fine-tuning procedure that achieves better adaption and improves classification performance in the target domain. Extensive experiments and ablation studies are conducted to demonstrate the effectiveness of the proposed method. Experimental results show that our proposed method consistently outperforms SOTA methods under various conditions.
AbstractList State of the art (SOTA) few-shot learning (FSL) methods suffer significant performance drop in the presence of domain differences between source and target datasets. The strong discrimination ability on the source dataset does not necessarily translate to high classification accuracy on the target dataset. In this work, we address this cross-domain few-shot learning (CDFSL) problem by boosting the generalization capability of the model. Specifically, we teach the model to capture broader variations of the feature distributions with a novel noise-enhanced supervised autoencoder (NSAE). NSAE trains the model by jointly reconstructing inputs and predicting the labels of inputs as well as their reconstructed pairs. Theoretical analysis based on intra-class correlation (ICC) shows that the feature embeddings learned from NSAE have stronger discrimination and generalization abilities in the target domain. We also take advantage of NSAE structure and propose a two-step fine-tuning procedure that achieves better adaption and improves classification performance in the target domain. Extensive experiments and ablation studies are conducted to demonstrate the effectiveness of the proposed method. Experimental results show that our proposed method consistently outperforms SOTA methods under various conditions.
Author Liang, Hanwen
Lu, Juwei
Zhang, Qiong
Dai, Peng
Author_xml – sequence: 1
  givenname: Hanwen
  surname: Liang
  fullname: Liang, Hanwen
  email: hanwen.liang@huawei.com
  organization: Huawei Noah's Ark Lab,Canada
– sequence: 2
  givenname: Qiong
  surname: Zhang
  fullname: Zhang, Qiong
  email: qiong.zhang@stat.ubc.ca
  organization: University of British Columbia,Department of Statistics,Vancouver,Canada
– sequence: 3
  givenname: Peng
  surname: Dai
  fullname: Dai, Peng
  email: peng.dai@huawei.com
  organization: Huawei Noah's Ark Lab,Canada
– sequence: 4
  givenname: Juwei
  surname: Lu
  fullname: Lu, Juwei
  email: juwei.lu@huawei.com
  organization: Huawei Noah's Ark Lab,Canada
BookMark eNotj9tOAjEYhKvRRECeQC_6Aot_D7vbXuIqSLLRCw-3pOz-KzXQkrZg8Oldojcz-SaZSWZILpx3SMgtgwljoO8WVfUhleZ8woGzCYDm-oyMdalYUeSSK8bzczLgQkFW5iCvyDDGLwChuSoGJNx7H5N1nzStkc7RYTAb-2OS9Y5WZmdWdmPTkdqego8xe_Bb08MMv7O49onWaII79Q_W0GdvI2bo1sY12NLX_Q7DoY9aOt0nj67xLYZrctmZTcTxv4_I--zxrXrK6pf5oprWmeUgUtYVedEpuYIGVYtiVXataplkLSsNQscb3V8zKBoFhSxOyqUSOQJKKUxnxIjc_O1aRFzugt2acFzqErTWpfgF2VpeyQ
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICCV48922.2021.00929
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 9781665428125
1665428120
EISSN 2380-7504
EndPage 9414
ExternalDocumentID 9709997
Genre orig-research
GroupedDBID 29O
6IE
6IF
6IH
6IK
6IL
6IM
6IN
AAJGR
AAWTH
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IPLJI
M43
OCL
RIE
RIL
RIO
RNS
ID FETCH-LOGICAL-i203t-f656f84b0ce8de3b7fd8d141d17ae0f2c9812ae3c80646c80624835e0e443afa3
IEDL.DBID RIE
ISICitedReferencesCount 60
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000798743208015&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 02:25:42 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-f656f84b0ce8de3b7fd8d141d17ae0f2c9812ae3c80646c80624835e0e443afa3
PageCount 11
ParticipantIDs ieee_primary_9709997
PublicationCentury 2000
PublicationDate 2021-Oct.
PublicationDateYYYYMMDD 2021-10-01
PublicationDate_xml – month: 10
  year: 2021
  text: 2021-Oct.
PublicationDecade 2020
PublicationTitle Proceedings / IEEE International Conference on Computer Vision
PublicationTitleAbbrev ICCV
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0039286
Score 2.4939816
Snippet State of the art (SOTA) few-shot learning (FSL) methods suffer significant performance drop in the presence of domain differences between source and target...
SourceID ieee
SourceType Publisher
StartPage 9404
SubjectTerms Boosting
Computer vision
Correlation
Feature extraction
Optimization and learning methods
Predictive models
Recognition and classification
Representation learning
Transfer learning
Transfer/Low-shot/Semi/Unsupervised Learning
Title Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder
URI https://ieeexplore.ieee.org/document/9709997
WOSCitedRecordID wos000798743208015&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/eLvHCXMwlV07T8MwGLRKxcBUoEW85YERt0nsxs4IgQqWqhIPdauc-DPNQFLlUcS_x05NERILS2RnieTXneK7-xC6okqA9lPP7DQFhKkxI4mfaLOWzfoSwKUE0Rab4NOpmM-jWQddb70wANCKz2Bom-1dvirSxv4qG0Xc8hm-g3Y4Dzdere9T18C8CJ01zvei0WMcvzIRBdZrFfhDGy0U_Sqg0uLHpPe_L--jwY8RD8-2EHOAOpAfop5jjtjty6qPytuiqKyAGRs-h12UtHNY4tjgYSuB_cSZ6VlYJHfFuzSdCXyQalnU2MWsvuF1JvG0yCogkC9bdQB-alb2QKlM86apC5t8qaAcoJfJ_XP8QFw1BZIFHq2JNsxNC5Z4KQgFNOFaCeUzX_lcgqeDNDJYL4GmwrCU0D4DZugZeMAYlVrSI9TNixyOEWZK60Sl0ViFhvBRkPYq34alaQHUH_MT1LdDuFhtAjMWbvRO_359hvbsHG0UcueoW5cNXKDddF1nVXnZzvIXMeCrgQ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NT8IwHG0QTfSECsZve_BoYR-FdUedEohISETDjXTrr7KDG9kHxv_edlSMiRcvS7vLkn69l_W990Po2hUMpB1ZaqcJIFR0KQntUKq1rNYXA49zYFWxCW88ZrOZP6mhm40XBgAq8Rm0dbO6yxdpVOpfZR3f03zG20LbunKWcWt9n7sK6FnPmONsy-8Mg-CVMt_RbivHbutwIf9XCZUKQfqN_317H7V-rHh4sgGZA1SD5BA1DHfEZmfmTZTdpWmuJcxYMTpswqSNxxIHChErEewnjlVPAyO5T9-56vThg-SLtMAmaPUNr2KOx2mcA4FkUekD8HO51EdKrpq3ZZHq7EsBWQu99B-mwYCYegokdiy3IFJxN8loaEXABLihJwUTNrWF7XGwpBP5Cu05uBFTPKWnnw5VBA0soNTlkrtHqJ6kCRwjTIWUoYj8rugpyucC15f5Oi5NMnDtrneCmnoI58t1ZMbcjN7p36-v0O5g-jSaj4bjxzO0p-drrZc7R_UiK-EC7USrIs6zy2rGvwBMaK7K
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%3Abook&rft.genre=proceeding&rft.title=Proceedings+%2F+IEEE+International+Conference+on+Computer+Vision&rft.atitle=Boosting+the+Generalization+Capability+in+Cross-Domain+Few-shot+Learning+via+Noise-enhanced+Supervised+Autoencoder&rft.au=Liang%2C+Hanwen&rft.au=Zhang%2C+Qiong&rft.au=Dai%2C+Peng&rft.au=Lu%2C+Juwei&rft.date=2021-10-01&rft.pub=IEEE&rft.eissn=2380-7504&rft.spage=9404&rft.epage=9414&rft_id=info:doi/10.1109%2FICCV48922.2021.00929&rft.externalDocID=9709997