{"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/every-untrue-label-is-untrue-in-its-own-way","title":"Every Untrue Label is Untrue in its Own Way: Controlling Error Type with the Log Bilinear Loss","arxiv_id":"1704.06062","date":"2017-04-20","proceeding":null,"authors":["Yehezkel S. Resheff","Amit Mandelbaum","Daphna Weinshall"],"abstract":"Deep learning has become the method of choice in many application domains of\nmachine learning in recent years, especially for multi-class classification\ntasks. The most common loss function used in this context is the cross-entropy\nloss, which reduces to the log loss in the typical case when there is a single\ncorrect response label. While this loss is insensitive to the identity of the\nassigned class in the case of misclassification, in practice it is often the\ncase that some errors may be more detrimental than others. Here we present the\nbilinear-loss (and related log-bilinear-loss) which differentially penalizes\nthe different wrong assignments of the model. We thoroughly test this method\nusing standard models and benchmark image datasets. As one application, we show\nthe ability of this method to better contain error within the correct\nsuper-class, in the hierarchically labeled CIFAR100 dataset, without affecting\nthe overall performance of the classifier.","url_abs":"http://arxiv.org/abs/1704.06062v1","url_pdf":"http://arxiv.org/pdf/1704.06062v1.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":"every-untrue-label-is-untrue-in-its-own-way","repo_url":"https://github.com/Hezi-Resheff/paper-log-bilinear-loss","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"multi-class-classification","task_name":"Multi-class Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}