Recent successes in self-supervised learning (SSL) model spatial co-occurrences of visual features either by masking portions of an image or by aggressively cropping it. Here, we propose a new way to model spatial co-occurrences by aligning local representations (before pooling) with a global image representation. We present CO-SSL, a family of instance discrimination methods and show that it outperforms previous methods on several datasets, including ImageNet-1K where it achieves 71.5% of Top-1 accuracy with 100 pre-training epochs. CO-SSL is also more robust to noise corruption, internal corruption, small adversarial attacks, and large training crop sizes. Our analysis further indicates that CO-SSL learns highly redundant local representations, which offers an explanation for its robustness. Overall, our work suggests that aligning local and global representations may be a powerful principle of unsupervised category learning.
Seeing SDG 6 from space: local-scale monitoring of piped water and sewage systems across Africa using satellite imagery and self-supervised learning
Mila – Quebec AI Institute(魁北克人工智能研究所);School of Computer Science, McGill University(麦吉尔大学计算机科学学院);Department of Earth and Environmental Engineering, Columbia University(哥伦比亚大学地球与环境工程系);Center for Learning the Earth with Artificial Intelligence and Physics (LEAP)(人工智能与物理学习地球中心(LEAP));Division of Natural and Applied Sciences, Duke Kunshan University(杜克-昆山大学自然科学与应用科学系)
Magnification-Aware Distillation (MAD): A Self-Supervised Framework for Unified Representation Learning in Gigapixel Whole-Slide Images
Center for Applied Artificial Intelligence, University of Kentucky(应用人工智能中心,肯塔基大学);Sanders-Brown Center on Aging, University of Kentucky(老龄化桑德斯-布朗中心,肯塔基大学);Department of Electrical and Computer Engineering, University of Kentucky(电气与计算机工程系,肯塔基大学);Division of Neuropathology, University of Pittsburgh(神经病理学部,匹兹堡大学);Department of Biomedical Informatics, Emory University(生物医学信息学系,埃默里大学);Department of Pathology and Laboratory Medicine, University of California Davis(病理学与实验室医学系,加州大学戴维斯分校);Department of Laboratory Medicine and Pathology, University of Texas Health(实验室医学与病理学系,德克萨斯大学健康科学中心)
Nonparametric Distribution Matching for Self-Supervised Whole-Slide Image Condensation
University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Washington State University(华盛顿州立大学);Hanoi Medical University(河内医科大学);VinUniversity(文森大学);Hanoi University of Science and Technology(河内科技大学)
Seeing in the Dark: A Teacher-Student Framework for Dark Video Action Recognition via Knowledge Distillation and Contrastive Learning
School of Information Technology, Monash University, Malaysia;Faculty of Information Technology, Monash University, Australia;Robotics \& Autonomous Systems Lab, TCS Research, India;Dept of Electrical Engineering, University of Malaya, Malaysia
Tracking by weakly-supervised learning and graph optimization for whole-embryo C. elegans lineages
Max-Delbrueck-Center for Molecular Medicine in the Helmholtz Association(亥姆霍兹联合会马克斯·德尔布吕克分子医学中心);Humboldt-Universität zu Berlin(柏林洪堡大学);Sloan Kettering Cancer Center(斯隆·凯特琳癌症中心);HHMI Janelia Research Campus(霍华德·休斯医学研究所珍妮莉亚研究园区)
Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images
Brigham and Women’s Hospital(布莱根妇女医院);Harvard Medical School(哈佛医学院);Broad Institute of Harvard and MIT(哈佛-麻省理工学院博德研究所);Dana-Farber Cancer Institute(达纳-法伯癌症研究所)
A weakly-supervised deep learning model for fast localisation and delineation of the skeleton, internal organs, and spinal canal on Whole-Body Diffusion-Weighted MRI (WB-DWI)
The Institute of Cancer Research(癌症研究所);The Royal Marsden NHS Foundation Trust(皇家马斯登NHS基金会信托);University Hospital Basel(巴塞尔大学医院)
Federated Learning for Computational Pathology on Gigapixel Whole Slide Images
Brigham and Women’s Hospital(布里格姆妇女医院);Harvard Medical School(哈佛医学院);Broad Institute of Harvard and MIT(哈佛-麻省理工学院布罗德研究所);Department of Pathology(病理系);Northwell Health(诺斯韦尔健康);Department of Biomedical Informatics(生物医学信息学系);Dana-Farber Cancer Institute(达纳-法伯癌症研究所);Cancer Program(癌症项目);Cancer Data Science(癌症数据科学)
Reconstructing neuronal anatomy from whole-brain images
Columbia University(哥伦比亚大学);Cold Spring Harbor Laboratory(冷泉港实验室);Princeton Neuroscience Institute(普林斯顿神经科学研究所);Massachusetts Institute of Technology(麻省理工学院);Allen Institute for Brain Science(艾伦脑科学研究所)
Weakly Supervised Arrhythmia Detection Based on Deep Convolutional Neural Network
School of Computer Science and Technology, Harbin Institute of Technology(哈尔滨工业大学计算机科学与技术学院);Peng Cheng Laboratory(鹏城实验室);School of Physics and Astronomy, University of Manchester(曼彻斯特大学物理与天文学院);Northeastern University(东北大学)