Sequential Rule Mining for Automated Design of Meta-heuristics

Meng, Weiyao; Qu, Rong

Abstract

With a recently defined AutoGCOP framework, the design of local search algorithms can be defined as the composition of the basic elementary algorithmic components. These compositions into the best algorithms thus retain useful knowledge of effective algorithm design. This paper investigates effective algorithmic compositions with sequential rule mining techniques to discover valuable knowledge in algorithm design. With the collected effective algorithmic compositions, sequential rules of basic algorithmic components are extracted and further analysed to automatically compose basic algorithmic components within the general AutoGCOP framework to develop new effective meta-heuristics. The sequential rules present superior performance in composing the basic algorithmic components for solving the benchmark vehicle routing problems with time window constraints, demonstrating its effectiveness in designing new algorithms automatically.

Citation

Meng, W., & Qu, R. (2023, July). Sequential Rule Mining for Automated Design of Meta-heuristics. Presented at GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion, New York, USA

• Presentation Conference Type: Edited Proceedings
• Conference Name: GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion
• Start Date: Jul 15, 2023
• End Date: Jul 20, 2023
• Acceptance Date: May 3, 2023
• Online Publication Date: Jul 24, 2023
• Publication Date: Jul 15, 2023
• Deposit Date: Jul 29, 2023
• Publicly Available Date: Aug 11, 2023
• Publisher: Association for Computing Machinery (ACM)
• Pages: 1727-1735
• Book Title: GECCO’23 Companion: Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion
• ISBN: 9798400701207
• DOI: https://doi.org/10.1145/3583133.3596303
• Keywords: Data Mining, Sequential Rule Mining, Automated Algorithm Design, Meta-heuristics, Vehicle Routing Problems
• Public URL: https://nottingham-repository.worktribe.com/output/23488788
• Publisher URL: https://dl.acm.org/doi/10.1145/3583133.3596303