{"language":[{"iso":"eng"}],"date_updated":"2022-08-24T12:45:39Z","type":"preprint","publication":"Machine Learning","author":[{"full_name":"Tornede, Alexander","id":"38209","last_name":"Tornede","first_name":"Alexander"},{"first_name":"Lukas","last_name":"Gehring","full_name":"Gehring, Lukas"},{"id":"40795","last_name":"Tornede","full_name":"Tornede, Tanja","first_name":"Tanja"},{"full_name":"Wever, Marcel Dominik","last_name":"Wever","id":"33176","first_name":"Marcel Dominik","orcid":" https://orcid.org/0000-0001-9782-6818"},{"id":"48129","last_name":"Hüllermeier","full_name":"Hüllermeier, Eyke","first_name":"Eyke"}],"abstract":[{"text":"The problem of selecting an algorithm that appears most suitable for a\r\nspecific instance of an algorithmic problem class, such as the Boolean\r\nsatisfiability problem, is called instance-specific algorithm selection. Over\r\nthe past decade, the problem has received considerable attention, resulting in\r\na number of different methods for algorithm selection. Although most of these\r\nmethods are based on machine learning, surprisingly little work has been done\r\non meta learning, that is, on taking advantage of the complementarity of\r\nexisting algorithm selection methods in order to combine them into a single\r\nsuperior algorithm selector. In this paper, we introduce the problem of meta\r\nalgorithm selection, which essentially asks for the best way to combine a given\r\nset of algorithm selectors. We present a general methodological framework for\r\nmeta algorithm selection as well as several concrete learning methods as\r\ninstantiations of this framework, essentially combining ideas of meta learning\r\nand ensemble learning. In an extensive experimental evaluation, we demonstrate\r\nthat ensembles of algorithm selectors can significantly outperform single\r\nalgorithm selectors and have the potential to form the new state of the art in\r\nalgorithm selection.","lang":"eng"}],"status":"public","project":[{"name":"SFB 901: SFB 901","_id":"1"},{"_id":"3","name":"SFB 901 - B: SFB 901 - Project Area B"},{"name":"SFB 901 - B2: SFB 901 - Subproject B2","_id":"10"}],"year":"2022","external_id":{"arxiv":["2107.09414"]},"user_id":"38209","citation":{"chicago":"Tornede, Alexander, Lukas Gehring, Tanja Tornede, Marcel Dominik Wever, and Eyke Hüllermeier. “Algorithm Selection on a Meta Level.” <i>Machine Learning</i>, 2022.","short":"A. Tornede, L. Gehring, T. Tornede, M.D. Wever, E. Hüllermeier, Machine Learning (2022).","bibtex":"@article{Tornede_Gehring_Tornede_Wever_Hüllermeier_2022, title={Algorithm Selection on a Meta Level}, journal={Machine Learning}, author={Tornede, Alexander and Gehring, Lukas and Tornede, Tanja and Wever, Marcel Dominik and Hüllermeier, Eyke}, year={2022} }","apa":"Tornede, A., Gehring, L., Tornede, T., Wever, M. D., &#38; Hüllermeier, E. (2022). Algorithm Selection on a Meta Level. In <i>Machine Learning</i>.","ieee":"A. Tornede, L. Gehring, T. Tornede, M. D. Wever, and E. Hüllermeier, “Algorithm Selection on a Meta Level,” <i>Machine Learning</i>. 2022.","ama":"Tornede A, Gehring L, Tornede T, Wever MD, Hüllermeier E. Algorithm Selection on a Meta Level. <i>Machine Learning</i>. Published online 2022.","mla":"Tornede, Alexander, et al. “Algorithm Selection on a Meta Level.” <i>Machine Learning</i>, 2022."},"date_created":"2022-04-12T11:55:18Z","title":"Algorithm Selection on a Meta Level","_id":"30865","department":[{"_id":"34"},{"_id":"7"},{"_id":"26"}]}