{"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/training-confidence-calibrated-classifiers","title":"Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples","arxiv_id":"1711.09325","date":"2017-11-26","proceeding":"ICLR 2018 1","authors":["Kimin Lee","Honglak Lee","Kibok Lee","Jinwoo Shin"],"abstract":"The problem of detecting whether a test sample is from in-distribution (i.e.,\ntraining distribution by a classifier) or out-of-distribution sufficiently\ndifferent from it arises in many real-world machine learning applications.\nHowever, the state-of-art deep neural networks are known to be highly\noverconfident in their predictions, i.e., do not distinguish in- and\nout-of-distributions. Recently, to handle this issue, several threshold-based\ndetectors have been proposed given pre-trained neural classifiers. However, the\nperformance of prior works highly depends on how to train the classifiers since\nthey only focus on improving inference procedures. In this paper, we develop a\nnovel training method for classifiers so that such inference algorithms can\nwork better. In particular, we suggest two additional terms added to the\noriginal loss (e.g., cross entropy). The first one forces samples from\nout-of-distribution less confident by the classifier and the second one is for\n(implicitly) generating most effective training samples for the first one. In\nessence, our method jointly trains both classification and generative neural\nnetworks for out-of-distribution. We demonstrate its effectiveness using deep\nconvolutional neural networks on various popular image datasets.","url_abs":"http://arxiv.org/abs/1711.09325v3","url_pdf":"http://arxiv.org/pdf/1711.09325v3.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":"training-confidence-calibrated-classifiers","repo_url":"https://github.com/alinlab/Confident_classifier","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"training-confidence-calibrated-classifiers","repo_url":"https://github.com/EpiSci/SafeAI","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"training-confidence-calibrated-classifiers","repo_url":"https://github.com/Paandaman/TRAINING-CONFIDENCE-CALIBRATED-CLASSIFIERS-FOR-DETECTING-OUT-OF-DISTRIBUTION-SAMPLES","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.09325","atlas_url":"https://app.syntology.ai/?focus=1711.09325","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}