{"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/confidence-from-invariance-to-image","title":"Confidence from Invariance to Image Transformations","arxiv_id":"1804.00657","date":"2018-04-02","proceeding":null,"authors":["Yuval Bahat","Gregory Shakhnarovich"],"abstract":"We develop a technique for automatically detecting the classification errors\nof a pre-trained visual classifier. Our method is agnostic to the form of the\nclassifier, requiring access only to classifier responses to a set of inputs.\nWe train a parametric binary classifier (error/correct) on a representation\nderived from a set of classifier responses generated from multiple copies of\nthe same input, each subject to a different natural image transformation. Thus,\nwe establish a measure of confidence in classifier's decision by analyzing the\ninvariance of its decision under various transformations. In experiments with\nmultiple data sets (STL-10,CIFAR-100,ImageNet) and classifiers, we demonstrate\nnew state of the art for the error detection task. In addition, we apply our\ntechnique to novelty detection scenarios, where we also demonstrate state of\nthe art results.","url_abs":"http://arxiv.org/abs/1804.00657v1","url_pdf":"http://arxiv.org/pdf/1804.00657v1.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":"confidence-from-invariance-to-image","repo_url":"https://github.com/YuvalBahat/Confidence_From_Invariance","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"novelty-detection","task_name":"Novelty Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1804.00657","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}