{"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/on-the-performance-of-differential-evolution","title":"On the Performance of Differential Evolution for Hyperparameter Tuning","arxiv_id":"1904.06960","date":"2019-04-15","proceeding":null,"authors":["Mischa Schmidt","Shahd Safarani","Julia Gastinger","Tobias Jacobs","Sebastien Nicolas","Anett Schülke"],"abstract":"Automated hyperparameter tuning aspires to facilitate the application of\nmachine learning for non-experts. In the literature, different optimization\napproaches are applied for that purpose. This paper investigates the\nperformance of Differential Evolution for tuning hyperparameters of supervised\nlearning algorithms for classification tasks. This empirical study involves a\nrange of different machine learning algorithms and datasets with various\ncharacteristics to compare the performance of Differential Evolution with\nSequential Model-based Algorithm Configuration (SMAC), a reference Bayesian\nOptimization approach. The results indicate that Differential Evolution\noutperforms SMAC for most datasets when tuning a given machine learning\nalgorithm - particularly when breaking ties in a first-to-report fashion. Only\nfor the tightest of computational budgets SMAC performs better. On small\ndatasets, Differential Evolution outperforms SMAC by 19% (37% after\ntie-breaking). In a second experiment across a range of representative datasets\ntaken from the literature, Differential Evolution scores 15% (23% after\ntie-breaking) more wins than SMAC.","url_abs":"http://arxiv.org/abs/1904.06960v1","url_pdf":"http://arxiv.org/pdf/1904.06960v1.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":"on-the-performance-of-differential-evolution","repo_url":"https://github.com/MLStruckmann/mutation-misery","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"bayesian-optimization","task_name":"Bayesian Optimization"},{"task_slug":"smac","task_name":"SMAC"},{"task_slug":"smac-1","task_name":"SMAC+"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}