Mining potentially explanatory patterns via partial solutions.

Catalano, Giancarlo Antonino Pasquale Ignazio; Brownlee, Alexander E.I.; Cairns, David; McCall, John; Ainslie, Russell

Abstract

We introduce Partial Solutions to improve the explainability of genetic algorithms for combinatorial optimization. Partial Solutions represent beneficial traits found by analyzing a population, and are presented to the user for explainability, but also provide an explicit model from which new solutions can be generated. We present an algorithm that assembles a collection of explanatory Partial Solutions chosen to strike a balance between simplicity, high fitness and atomicity, that are shown to be able to solve standard optimization benchmarks.

Citation

CATALANO, G.A.P.I., BROWNLEE, A.E.I., CAIRNS, D., MCCALL, J. and AINSLIE, R. 2024. Mining potentially explanatory patterns via partial solutions. In GECCO'24 companion: proceedings of the 2024 Genetic and evolutionary computation conference companion 2024 (GECCO'24 companion), 14-18 July 2024, Melbourne, Australia. New York: ACM [online], pages 567-570. Available from: https://doi.org/10.1145/3638530.3654318

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2024 Genetic and evolutionary computation conference companion 2024 (GECCO'24 companion)
• Start Date: Jul 14, 2024
• End Date: Jul 18, 2024
• Acceptance Date: Mar 21, 2024
• Online Publication Date: Jul 14, 2024
• Publication Date: Aug 1, 2024
• Deposit Date: Sep 5, 2024
• Publicly Available Date: Sep 5, 2024
• Publisher: Association for Computing Machinery (ACM)
• Peer Reviewed: Peer Reviewed
• Pages: 567-570
• Book Title: GECCO'24 companion: proceedings of the 2024 Genetic and evolutionary computation conference companion 2024 (GECCO'24 companion)
• ISBN: 9798400704956
• DOI: https://doi.org/10.1145/3638530.3654318
• Keywords: Genetic algorithms; Explainable AI (XAI); Combinatorial optimization problems
• Public URL: https://rgu-repository.worktribe.com/output/2423381