{"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/dimensionalitys-blessing-clustering-images-by","title":"Dimensionality's Blessing: Clustering Images by Underlying Distribution","arxiv_id":"1804.02624","date":"2018-04-08","proceeding":"CVPR 2018 6","authors":["Wen-Yan Lin","Siying Liu","Jian-Huang Lai","Yasuyuki Matsushita"],"abstract":"Many high dimensional vector distances tend to a constant. This is typically\nconsidered a negative \"contrast-loss\" phenomenon that hinders clustering and\nother machine learning techniques. We reinterpret \"contrast-loss\" as a\nblessing. Re-deriving \"contrast-loss\" using the law of large numbers, we show\nit results in a distribution's instances concentrating on a thin \"hyper-shell\".\nThe hollow center means apparently chaotically overlapping distributions are\nactually intrinsically separable. We use this to develop\ndistribution-clustering, an elegant algorithm for grouping of data points by\ntheir (unknown) underlying distribution. Distribution-clustering, creates\nnotably clean clusters from raw unlabeled data, estimates the number of\nclusters for itself and is inherently robust to \"outliers\" which form their own\nclusters. This enables trawling for patterns in unorganized data and may be the\nkey to enabling machine intelligence.","url_abs":"http://arxiv.org/abs/1804.02624v1","url_pdf":"http://arxiv.org/pdf/1804.02624v1.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":"dimensionalitys-blessing-clustering-images-by","repo_url":"https://github.com/EricElmoznino/distribution_clustering","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}