{"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-from-complementary-labels","title":"Learning from Complementary Labels","arxiv_id":"1705.07541","date":"2017-05-22","proceeding":"NeurIPS 2017 12","authors":["Takashi Ishida","Gang Niu","Weihua Hu","Masashi Sugiyama"],"abstract":"Collecting labeled data is costly and thus a critical bottleneck in\nreal-world classification tasks. To mitigate this problem, we propose a novel\nsetting, namely learning from complementary labels for multi-class\nclassification. A complementary label specifies a class that a pattern does not\nbelong to. Collecting complementary labels would be less laborious than\ncollecting ordinary labels, since users do not have to carefully choose the\ncorrect class from a long list of candidate classes. However, complementary\nlabels are less informative than ordinary labels and thus a suitable approach\nis needed to better learn from them. In this paper, we show that an unbiased\nestimator to the classification risk can be obtained only from complementarily\nlabeled data, if a loss function satisfies a particular symmetric condition. We\nderive estimation error bounds for the proposed method and prove that the\noptimal parametric convergence rate is achieved. We further show that learning\nfrom complementary labels can be easily combined with learning from ordinary\nlabels (i.e., ordinary supervised learning), providing a highly practical\nimplementation of the proposed method. Finally, we experimentally demonstrate\nthe usefulness of the proposed methods.","url_abs":"http://arxiv.org/abs/1705.07541v2","url_pdf":"http://arxiv.org/pdf/1705.07541v2.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-from-complementary-labels","repo_url":"https://github.com/takashiishida/comp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"multi-class-classification","task_name":"Multi-class Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.07541","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}