Domain adaptation and representation transfer, and distributed and collaborative learning : second MICCAI Workshop, DART 2020, and first MICCAI Workshop, DCL 2020, held in conjunction with MICCAI 2020, Lima, Peru, October 4-8, 2020, Pro...最新文献

筛选

英文中文

Parts2Whole: Self-supervised Contrastive Learning via Reconstruction. Parts2Whole：通过重构进行自监督对比学习

Domain adaptation and representation transfer, and distributed and collaborative learning : second MICCAI Workshop, DART 2020, and first MICCAI Workshop, DCL 2020, held in conjunction with MICCAI 2020, Lima, Peru, October 4-8, 2020, Pro... Pub Date : 2020-10-01 Epub Date: 2020-09-26 DOI: 10.1007/978-3-030-60548-3_9

Ruibin Feng, Zongwei Zhou, Michael B Gotway, Jianming Liang

{"title":"Parts2Whole: Self-supervised Contrastive Learning via Reconstruction.","authors":"Ruibin Feng, Zongwei Zhou, Michael B Gotway, Jianming Liang","doi":"10.1007/978-3-030-60548-3_9","DOIUrl":"10.1007/978-3-030-60548-3_9","url":null,"abstract":"Contrastive representation learning is the state of the art in computer vision, but requires huge mini-batch sizes, special network design, or memory banks, making it unappealing for 3D medical imaging, while in 3D medical imaging, reconstruction-based self-supervised learning reaches a new height in performance, but lacks mechanisms to learn contrastive representation; therefore, this paper proposes a new framework for self-supervised contrastive learning via reconstruction, called Parts2Whole, because it exploits the universal and intrinsic part-whole relationship to learn contrastive representation without using contrastive loss: Reconstructing an image (whole) from its own parts compels the model to learn similar latent features for all its own parts, while reconstructing different images (wholes) from their respective parts forces the model to simultaneously push those parts belonging to different wholes farther apart from each other in the latent space; thereby the trained model is capable of distinguishing images. We have evaluated our Parts2Whole on five distinct imaging tasks covering both classification and segmentation, and compared it with four competing publicly available 3D pretrained models, showing that Parts2Whole significantly outperforms in two out of five tasks while achieves competitive performance on the rest three. This superior performance is attributable to the contrastive representations learned with Parts2Whole. Codes and pretrained models are available at github.com/JLiangLab/Parts2Whole.","PeriodicalId":93545,"journal":{"name":"Domain adaptation and representation transfer, and distributed and collaborative learning : second MICCAI Workshop, DART 2020, and first MICCAI Workshop, DCL 2020, held in conjunction with MICCAI 2020, Lima, Peru, October 4-8, 2020, Pro...","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9197952/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86248628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Domain Adaptation and Representation Transfer, and Distributed and Collaborative Learning: Second MICCAI Workshop, DART 2020, and First MICCAI Workshop, DCL 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4–8, 2020, Proceedings 领域适应和表征转移，以及分布式和协作学习:第二届MICCAI研讨会，DART 2020，第一届MICCAI研讨会，DCL 2020，与MICCAI 2020一起举行，秘鲁利马，2020年10月4日至8日，会议记录

Shadi Albarqouni, S. Bakas, K. Kamnitsas

引用次数: 11