001037115 001__ 1037115
001037115 005__ 20250203103155.0
001037115 0247_ $$2datacite_doi$$a10.34734/FZJ-2025-00462
001037115 037__ $$aFZJ-2025-00462
001037115 1001_ $$0P:(DE-Juel1)190396$$aWang, Qin$$b0$$eCorresponding author$$ufzj
001037115 1112_ $$aThe Thirty-Eighth Annual Conference on Neural Information Processing Systems Workshop: Self-Supervised Learning - Theory and Practice$$cVancouver$$d2024-12-10 - 2024-12-15$$gNeurIPS 2024$$wCanada
001037115 245__ $$aEquivariant Representation Learning for Augmentation-based Self-Supervised Learning  via Image Reconstruction
001037115 260__ $$c2024
001037115 3367_ $$033$$2EndNote$$aConference Paper
001037115 3367_ $$2BibTeX$$aINPROCEEDINGS
001037115 3367_ $$2DRIVER$$aconferenceObject
001037115 3367_ $$2ORCID$$aCONFERENCE_POSTER
001037115 3367_ $$2DataCite$$aOutput Types/Conference Poster
001037115 3367_ $$0PUB:(DE-HGF)24$$2PUB:(DE-HGF)$$aPoster$$bposter$$mposter$$s1736780632_25079$$xAfter Call
001037115 520__ $$aAugmentation-based self-supervised learning methods have shown remarkable success in self-supervised visual representation learning, excelling in learning invariant features but often neglecting equivariant ones. This limitation reduces the generalizability of foundation models, particularly for downstream tasks requiring equivariance. We propose integrating an image reconstruction task as an auxiliary component in augmentation-based self-supervised learning algorithms to facilitate equivariant feature learning without additional parameters. Our method implements a cross-attention mechanism to blend features learned from two augmented views, subsequently reconstructing one of them. This approach is adaptable to various datasets and augmented-pair based learning methods. We evaluate its effectiveness on learning equivariant features through multiple linear regression tasks and downstream applications on both artificial (3DIEBench) and natural (ImageNet) datasets. Results consistently demonstrate significant improvements over standard augmentation-based self-supervised learning methods and state-of-the-art approaches, particularly excelling in scenarios involving combined augmentations. Our method enhances the learning of both invariant and equivariant features, leading to more robust and generalizable visual representations for computer vision tasks.
001037115 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001037115 536__ $$0G:(DE-HGF)POF4-5112$$a5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x1
001037115 536__ $$0G:(DE-Juel1)Helmholtz-SLNS$$aSLNS - SimLab Neuroscience (Helmholtz-SLNS)$$cHelmholtz-SLNS$$x2
001037115 7001_ $$0P:(DE-Juel1)129347$$aKrajsek, Kai$$b1$$ufzj
001037115 7001_ $$0P:(DE-Juel1)129394$$aScharr, Hanno$$b2$$ufzj
001037115 8564_ $$uhttps://juser.fz-juelich.de/record/1037115/files/poster.pdf$$yOpenAccess
001037115 909CO $$ooai:juser.fz-juelich.de:1037115$$pdriver$$pVDB$$popen_access$$popenaire
001037115 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)190396$$aForschungszentrum Jülich$$b0$$kFZJ
001037115 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)129347$$aForschungszentrum Jülich$$b1$$kFZJ
001037115 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)129394$$aForschungszentrum Jülich$$b2$$kFZJ
001037115 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001037115 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5112$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x1
001037115 9141_ $$y2024
001037115 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001037115 9201_ $$0I:(DE-Juel1)IAS-8-20210421$$kIAS-8$$lDatenanalyse und Maschinenlernen$$x0
001037115 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x1
001037115 9801_ $$aFullTexts
001037115 980__ $$aposter
001037115 980__ $$aVDB
001037115 980__ $$aUNRESTRICTED
001037115 980__ $$aI:(DE-Juel1)IAS-8-20210421
001037115 980__ $$aI:(DE-Juel1)JSC-20090406