{"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/unsupervised-learning-by-predicting-noise","title":"Unsupervised Learning by Predicting Noise","arxiv_id":"1704.05310","date":"2017-04-18","proceeding":"ICML 2017 8","authors":["Piotr Bojanowski","Armand Joulin"],"abstract":"Convolutional neural networks provide visual features that perform remarkably\nwell in many computer vision applications. However, training these networks\nrequires significant amounts of supervision. This paper introduces a generic\nframework to train deep networks, end-to-end, with no supervision. We propose\nto fix a set of target representations, called Noise As Targets (NAT), and to\nconstrain the deep features to align to them. This domain agnostic approach\navoids the standard unsupervised learning issues of trivial solutions and\ncollapsing of features. Thanks to a stochastic batch reassignment strategy and\na separable square loss function, it scales to millions of images. The proposed\napproach produces representations that perform on par with state-of-the-art\nunsupervised methods on ImageNet and Pascal VOC.","url_abs":"http://arxiv.org/abs/1704.05310v1","url_pdf":"http://arxiv.org/pdf/1704.05310v1.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":"unsupervised-learning-by-predicting-noise","repo_url":"https://github.com/facebookresearch/noise-as-targets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.05310","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}