{"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/spline-based-probability-calibration","title":"Spline-Based Probability Calibration","arxiv_id":"1809.07751","date":"2018-09-20","proceeding":null,"authors":["Brian Lucena"],"abstract":"In many classification problems it is desirable to output well-calibrated\nprobabilities on the different classes. We propose a robust, non-parametric\nmethod of calibrating probabilities called SplineCalib that utilizes smoothing\nsplines to determine a calibration function. We demonstrate how applying\ncertain transformations as part of the calibration process can improve\nperformance on problems in deep learning and other domains where the scores\ntend to be \"overconfident\". We adapt the approach to multi-class problems and\nfind that better calibration can improve accuracy as well as log-loss by better\nresolving uncertain cases. Finally, we present a cross-validated approach to\ncalibration which conserves data. Significant improvements to log-loss and\naccuracy are shown on several different problems. We also introduce the\nml-insights python package which contains an implementation of the SplineCalib\nalgorithm.","url_abs":"http://arxiv.org/abs/1809.07751v1","url_pdf":"http://arxiv.org/pdf/1809.07751v1.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":"spline-based-probability-calibration","repo_url":"https://github.com/numeristical/introspective","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.07751","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}