%0 Conference Paper %1 Knauer2023-im %A Knauer, Ricardo %A Rodner, Erik %B KI 2023: Advances in Artificial Intelligence: 46th German Conference on AI, Berlin, Germany, September 26--29, 2023, Proceedings %C Berlin, Heidelberg %D 2023 %I Springer-Verlag %K Zno best conic cost-sensitive interpretable learning machine meta-learning mixed-integer optimization selection subset %P 114--129 %T Cost-Sensitive Best Subset Selection for Logistic Regression: A Mixed-Integer Conic Optimization Perspective %X A key challenge in machine learning is to design interpretable models that can reduce their inputs to the best subset for making transparent predictions, especially in the clinical domain. In this work, we propose a certifiably optimal feature selection procedure for logistic regression from a mixed-integer conic optimization perspective that can take an auxiliary cost to obtain features into account. Based on an extensive review of the literature, we carefully create a synthetic dataset generator for clinical prognostic model research. This allows us to systematically evaluate different heuristic and optimal cardinality- and budget-constrained feature selection procedures. The analysis shows key limitations of the methods for the low-data regime and when confronted with label noise. Our paper not only provides empirical recommendations for suitable methods and dataset designs, but also paves the way for future research in the area of meta-learning.

@inproceedings{Knauer2023-im, abstract = {A key challenge in machine learning is to design interpretable models that can reduce their inputs to the best subset for making transparent predictions, especially in the clinical domain. In this work, we propose a certifiably optimal feature selection procedure for logistic regression from a mixed-integer conic optimization perspective that can take an auxiliary cost to obtain features into account. Based on an extensive review of the literature, we carefully create a synthetic dataset generator for clinical prognostic model research. This allows us to systematically evaluate different heuristic and optimal cardinality- and budget-constrained feature selection procedures. The analysis shows key limitations of the methods for the low-data regime and when confronted with label noise. Our paper not only provides empirical recommendations for suitable methods and dataset designs, but also paves the way for future research in the area of meta-learning.}, added-at = {2025-01-07T13:17:46.000+0100}, address = {Berlin, Heidelberg}, author = {Knauer, Ricardo and Rodner, Erik}, biburl = {https://puma.scadsai.uni-leipzig.de/bibtex/24df2b2b5a46bbe20000e9dc9f92d9ad0/scadsfct}, booktitle = {{KI} 2023: Advances in Artificial Intelligence: 46th German Conference on {AI}, Berlin, Germany, September 26--29, 2023, Proceedings}, interhash = {88b4f129bad2e850d26d0763d2ef31bd}, intrahash = {4df2b2b5a46bbe20000e9dc9f92d9ad0}, keywords = {Zno best conic cost-sensitive interpretable learning machine meta-learning mixed-integer optimization selection subset}, location = {Berlin, Germany}, pages = {114--129}, publisher = {Springer-Verlag}, timestamp = {2025-02-04T10:56:14.000+0100}, title = {{Cost-Sensitive} Best Subset Selection for Logistic Regression: A {Mixed-Integer} Conic Optimization Perspective}, year = 2023 }