{"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/learning-a-metric-for-class-conditional-knn","title":"Learning a metric for class-conditional KNN","arxiv_id":"1607.03050","date":"2016-07-11","proceeding":null,"authors":["Daniel Jiwoong Im","Graham W. Taylor"],"abstract":"Naive Bayes Nearest Neighbour (NBNN) is a simple and effective framework\nwhich addresses many of the pitfalls of K-Nearest Neighbour (KNN)\nclassification. It has yielded competitive results on several computer vision\nbenchmarks. Its central tenet is that during NN search, a query is not compared\nto every example in a database, ignoring class information. Instead, NN\nsearches are performed within each class, generating a score per class. A key\nproblem with NN techniques, including NBNN, is that they fail when the data\nrepresentation does not capture perceptual (e.g.~class-based) similarity. NBNN\ncircumvents this by using independent engineered descriptors (e.g.~SIFT). To\nextend its applicability outside of image-based domains, we propose to learn a\nmetric which captures perceptual similarity. Similar to how Neighbourhood\nComponents Analysis optimizes a differentiable form of KNN classification, we\npropose \"Class Conditional\" metric learning (CCML), which optimizes a soft form\nof the NBNN selection rule. Typical metric learning algorithms learn either a\nglobal or local metric. However, our proposed method can be adjusted to a\nparticular level of locality by tuning a single parameter. An empirical\nevaluation on classification and retrieval tasks demonstrates that our proposed\nmethod clearly outperforms existing learned distance metrics across a variety\nof image and non-image datasets.","url_abs":"http://arxiv.org/abs/1607.03050v1","url_pdf":"http://arxiv.org/pdf/1607.03050v1.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":"learning-a-metric-for-class-conditional-knn","repo_url":"https://github.com/jiwoongim/CCML","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}