{"date_created":"2023-08-04T07:37:30Z","doi":"10.1109/SSCI47803.2020.9308510","_id":"46328","status":"public","year":"2020","author":[{"first_name":"Raphael Patrick","full_name":"Prager, Raphael Patrick","last_name":"Prager"},{"orcid":"0000-0002-9788-8282","id":"100740","first_name":"Heike","full_name":"Trautmann, Heike","last_name":"Trautmann"},{"first_name":"Hao","last_name":"Wang","full_name":"Wang, Hao"},{"first_name":"Thomas H. W.","full_name":"Bäck, Thomas H. W.","last_name":"Bäck"},{"full_name":"Kerschke, Pascal","last_name":"Kerschke","first_name":"Pascal"}],"user_id":"15504","type":"conference","citation":{"apa":"Prager, R. P., Trautmann, H., Wang, H., Bäck, T. H. W., &#38; Kerschke, P. (2020). Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis. <i>Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)</i>, 996–1003. <a href=\"https://doi.org/10.1109/SSCI47803.2020.9308510\">https://doi.org/10.1109/SSCI47803.2020.9308510</a>","mla":"Prager, Raphael Patrick, et al. “Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis.” <i>Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)</i>, 2020, pp. 996–1003, doi:<a href=\"https://doi.org/10.1109/SSCI47803.2020.9308510\">10.1109/SSCI47803.2020.9308510</a>.","chicago":"Prager, Raphael Patrick, Heike Trautmann, Hao Wang, Thomas H. W. Bäck, and Pascal Kerschke. “Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis.” In <i>Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)</i>, 996–1003. Canberra, Australia, 2020. <a href=\"https://doi.org/10.1109/SSCI47803.2020.9308510\">https://doi.org/10.1109/SSCI47803.2020.9308510</a>.","ama":"Prager RP, Trautmann H, Wang H, Bäck THW, Kerschke P. Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis. In: <i>Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)</i>. ; 2020:996–1003. doi:<a href=\"https://doi.org/10.1109/SSCI47803.2020.9308510\">10.1109/SSCI47803.2020.9308510</a>","bibtex":"@inproceedings{Prager_Trautmann_Wang_Bäck_Kerschke_2020, place={Canberra, Australia}, title={Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis}, DOI={<a href=\"https://doi.org/10.1109/SSCI47803.2020.9308510\">10.1109/SSCI47803.2020.9308510</a>}, booktitle={Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)}, author={Prager, Raphael Patrick and Trautmann, Heike and Wang, Hao and Bäck, Thomas H. W. and Kerschke, Pascal}, year={2020}, pages={996–1003} }","ieee":"R. P. Prager, H. Trautmann, H. Wang, T. H. W. Bäck, and P. Kerschke, “Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis,” in <i>Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)</i>, 2020, pp. 996–1003, doi: <a href=\"https://doi.org/10.1109/SSCI47803.2020.9308510\">10.1109/SSCI47803.2020.9308510</a>.","short":"R.P. Prager, H. Trautmann, H. Wang, T.H.W. Bäck, P. Kerschke, in: Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, Australia, 2020, pp. 996–1003."},"department":[{"_id":"34"},{"_id":"819"}],"language":[{"iso":"eng"}],"place":"Canberra, Australia","page":"996–1003","date_updated":"2023-10-16T13:04:15Z","title":"Per-Instance Configuration of the Modularized CMA-ES by Means of Classifier Chains and Exploratory Landscape Analysis","publication":"Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI)","abstract":[{"text":"In this paper, we rely on previous work proposing a modularized version of CMA-ES, which captures several alterations to the conventional CMA-ES developed in recent years. Each alteration provides significant advantages under certain problem properties, e.g., multi-modality, high conditioning. These distinct advancements are implemented as modules which result in 4608 unique versions of CMA-ES. Previous findings illustrate the competitive advantage of enabling and disabling the aforementioned modules for different optimization problems. Yet, this modular CMA-ES is lacking a method to automatically determine when the activation of specific modules is auspicious and when it is not. We propose a well-performing instance-specific algorithm configuration model which selects an (almost) optimal configuration of modules for a given problem instance. In addition, the structure of this configuration model is able to capture inter-dependencies between modules, e.g., two (or more) modules might only be advantageous in unison for some problem types, making the orchestration of modules a crucial task. This is accomplished by chaining multiple random forest classifiers together into a so-called Classifier Chain based on a set of numerical features extracted by means of Exploratory Landscape Analysis (ELA) to describe the given problem instances.","lang":"eng"}]}