{"publication":"Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence","date_created":"2024-08-19T12:43:55Z","doi":"10.24963/ijcai.2024/479","publisher":"International Joint Conferences on Artificial Intelligence Organization","user_id":"87189","title":"ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling","citation":{"chicago":"Kouagou, N’Dah Jean, Stefan Heindorf, Caglar Demir, and Axel-Cyrille Ngonga Ngomo. “ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling.” In Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence Organization, 2024. https://doi.org/10.24963/ijcai.2024/479.","bibtex":"@inproceedings{Kouagou_Heindorf_Demir_Ngonga Ngomo_2024, title={ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling}, DOI={10.24963/ijcai.2024/479}, booktitle={Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence}, publisher={International Joint Conferences on Artificial Intelligence Organization}, author={Kouagou, N’Dah Jean and Heindorf, Stefan and Demir, Caglar and Ngonga Ngomo, Axel-Cyrille}, year={2024} }","mla":"Kouagou, N’Dah Jean, et al. “ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling.” Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence, International Joint Conferences on Artificial Intelligence Organization, 2024, doi:10.24963/ijcai.2024/479.","short":"N.J. Kouagou, S. Heindorf, C. Demir, A.-C. Ngonga Ngomo, in: Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence, International Joint Conferences on Artificial Intelligence Organization, 2024.","apa":"Kouagou, N. J., Heindorf, S., Demir, C., & Ngonga Ngomo, A.-C. (2024). ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling. Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence. https://doi.org/10.24963/ijcai.2024/479","ieee":"N. J. Kouagou, S. Heindorf, C. Demir, and A.-C. Ngonga Ngomo, “ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling,” 2024, doi: 10.24963/ijcai.2024/479.","ama":"Kouagou NJ, Heindorf S, Demir C, Ngonga Ngomo A-C. ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling. In: Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence Organization; 2024. doi:10.24963/ijcai.2024/479"},"publication_status":"published","date_updated":"2024-08-19T12:46:04Z","language":[{"iso":"eng"}],"author":[{"first_name":"N'Dah Jean","last_name":"Kouagou","full_name":"Kouagou, N'Dah Jean"},{"full_name":"Heindorf, Stefan","first_name":"Stefan","last_name":"Heindorf"},{"full_name":"Demir, Caglar","last_name":"Demir","first_name":"Caglar"},{"first_name":"Axel-Cyrille","last_name":"Ngonga Ngomo","full_name":"Ngonga Ngomo, Axel-Cyrille"}],"_id":"55653","type":"conference","year":"2024","status":"public","abstract":[{"lang":"eng","text":"We consider the problem of class expression learning using cardinality-minimal sets of examples. Recent class expression learning approaches employ deep neural networks and have demonstrated tremendous performance improvements in execution time and quality of the computed solutions. However, they lack generalization capabilities when it comes to the number of examples used in a learning problem, i.e., they often perform poorly on unseen learning problems where only a few examples are given. In this work, we propose a generalization of the classical class expression learning problem to address the limitations above. In short, our generalized learning problem (GLP) forces learning systems to solve the classical class expression learning problem using the smallest possible subsets of examples, thereby improving the learning systems' ability to solve unseen learning problems with arbitrary numbers of examples. Moreover, we develop ROCES, a learning algorithm for synthesis-based approaches to solve GLP. Experimental results suggest that post training, ROCES outperforms existing synthesis-based approaches on out-of-distribution learning problems while remaining highly competitive overall."}]}