{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/local-aggregation-for-unsupervised-learning","title":"Local Aggregation for Unsupervised Learning of Visual Embeddings","arxiv_id":"1903.12355","date":"2019-03-29","proceeding":"ICCV 2019 10","authors":["Chengxu Zhuang","Alex Lin Zhai","Daniel Yamins"],"abstract":"Unsupervised approaches to learning in neural networks are of substantial\ninterest for furthering artificial intelligence, both because they would enable\nthe training of networks without the need for large numbers of expensive\nannotations, and because they would be better models of the kind of\ngeneral-purpose learning deployed by humans. However, unsupervised networks\nhave long lagged behind the performance of their supervised counterparts,\nespecially in the domain of large-scale visual recognition. Recent developments\nin training deep convolutional embeddings to maximize non-parametric instance\nseparation and clustering objectives have shown promise in closing this gap.\nHere, we describe a method that trains an embedding function to maximize a\nmetric of local aggregation, causing similar data instances to move together in\nthe embedding space, while allowing dissimilar instances to separate. This\naggregation metric is dynamic, allowing soft clusters of different scales to\nemerge. We evaluate our procedure on several large-scale visual recognition\ndatasets, achieving state-of-the-art unsupervised transfer learning performance\non object recognition in ImageNet, scene recognition in Places 205, and object\ndetection in PASCAL VOC.","url_abs":"http://arxiv.org/abs/1903.12355v2","url_pdf":"http://arxiv.org/pdf/1903.12355v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"local-aggregation-for-unsupervised-learning","repo_url":"https://github.com/neuroailab/LocalAggregation-Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"contrastive-learning","task_name":"Contrastive Learning"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-recognition","task_name":"Object Recognition"},{"task_slug":"scene-recognition","task_name":"Scene Recognition"},{"task_slug":"self-supervised-image-classification","task_name":"Self-Supervised Image Classification"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/contrastive-learning-on-imagenet-1k","task":"Contrastive Learning","dataset":"imagenet-1k","model":"ResNet50","rank_in_archive_order":12,"of":14,"metrics":{"ImageNet Top-1 Accuracy":"60.2"},"uses_additional_data":false},{"leaderboard":"/sota/self-supervised-image-classification-on","task":"Self-Supervised Image Classification","dataset":"ImageNet","model":"LocalAgg (ResNet-50)","rank_in_archive_order":126,"of":144,"metrics":{"Number of Params":"24M","Top 1 Accuracy":"60.2%"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1903.12355","atlas_url":"https://app.syntology.ai/?focus=1903.12355","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}