{"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/tunability-importance-of-hyperparameters-of","title":"Tunability: Importance of Hyperparameters of Machine Learning Algorithms","arxiv_id":"1802.09596","date":"2018-02-26","proceeding":null,"authors":["Philipp Probst","Bernd Bischl","Anne-Laure Boulesteix"],"abstract":"Modern supervised machine learning algorithms involve hyperparameters that\nhave to be set before running them. Options for setting hyperparameters are\ndefault values from the software package, manual configuration by the user or\nconfiguring them for optimal predictive performance by a tuning procedure. The\ngoal of this paper is two-fold. Firstly, we formalize the problem of tuning\nfrom a statistical point of view, define data-based defaults and suggest\ngeneral measures quantifying the tunability of hyperparameters of algorithms.\nSecondly, we conduct a large-scale benchmarking study based on 38 datasets from\nthe OpenML platform and six common machine learning algorithms. We apply our\nmeasures to assess the tunability of their parameters. Our results yield\ndefault values for hyperparameters and enable users to decide whether it is\nworth conducting a possibly time consuming tuning strategy, to focus on the\nmost important hyperparameters and to chose adequate hyperparameter spaces for\ntuning.","url_abs":"http://arxiv.org/abs/1802.09596v3","url_pdf":"http://arxiv.org/pdf/1802.09596v3.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":"tunability-importance-of-hyperparameters-of","repo_url":"https://github.com/PhilippPro/tunability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"tunability-importance-of-hyperparameters-of","repo_url":"https://github.com/mi2-warsaw/MI2DataLab_Seminarium","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"benchmarking","task_name":"Benchmarking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.09596","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}