Print Email Facebook Twitter Symbolic regression and feature construction with GP-GOMEA applied to radiotherapy dose reconstruction of childhood cancer survivors Title Symbolic regression and feature construction with GP-GOMEA applied to radiotherapy dose reconstruction of childhood cancer survivors Author Virgolin, M. (TU Delft Algorithmics; Centrum Wiskunde & Informatica (CWI)) Alderliesten, Tanja (Amsterdam UMC) Bel, Arjan (Amsterdam UMC) Witteveen, C. (TU Delft Algorithmics) Bosman, P.A.N. (TU Delft Algorithmics; Centrum Wiskunde & Informatica (CWI)) Date 2018 Abstract The recently introduced Gene-pool Optimal Mixing Evolutionary Algorithm for Genetic Programming (GP-GOMEA) has been shown to find much smaller solutions of equally high quality compared to other state-of-the-art GP approaches. This is an interesting aspect as small solutions better enable human interpretation. In this paper, an adaptation of GP-GOMEA to tackle real-world symbolic regression is proposed, in order to find small yet accurate mathematical expressions, and with an application to a problem of clinical interest. For radiotherapy dose reconstruction, a model is sought that captures anatomical patient similarity. This problem is particularly interesting because while features are patient-specific, the variable to regress is a distance, and is defined over patient pairs. We show that on benchmark problems as well as on the application, GP-GOMEA outperforms variants of standard GP. To find even more accurate models, we further consider an evolutionary meta learning approach, where GP-GOMEA is used to construct small, yet effective features for a different machine learning algorithm. Experimental results show how this approach significantly improves the performance of linear regression, support vector machines, and random forest, while providing meaningful and interpretable features. Subject Dose reconstructionFeature constructionGenetic programmingGOMEAMachine learningRadiotherapy To reference this document use: http://resolver.tudelft.nl/uuid:0953c693-0042-49da-8bb5-522158c1034f DOI https://doi.org/10.1145/3205455.3205604 Publisher Association for Computing Machinery (ACM), New York, NY ISBN 978-1-4503-5618-3 Source Proceedings of the 2018 Genetic and Evolutionary Computation Conference, GECCO 2018 Event GECCO 2018, 2018-07-15 → 2018-07-19, Kyoto, Japan Part of collection Institutional Repository Document type conference paper Rights © 2018 M. Virgolin, Tanja Alderliesten, Arjan Bel, C. Witteveen, P.A.N. Bosman Files PDF 46824662_kika_gecco_2018_2_.pdf 4.54 MB Close viewer /islandora/object/uuid:0953c693-0042-49da-8bb5-522158c1034f/datastream/OBJ/view