UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Measuring failed disruption propagation in genetic programming

Langdon, WB; Al-Subaihin, A; Clark, D; (2022) Measuring failed disruption propagation in genetic programming. In: GECCO '22: Proceedings of the Genetic and Evolutionary Computation Conference. (pp. pp. 964-972). ACM: New York, NY, United States. Green open access

[thumbnail of langdon_2022_GECCO2.pdf]
Preview
PDF
langdon_2022_GECCO2.pdf - Other

Download (1MB) | Preview

Abstract

Information theory explains the robustness of deep GP trees, with on average up to 83.3% of crossover run time disruptions failing to propagate to the root node, and so having no impact on fitness, leading to phenotypic convergence. Monte Carlo simulations of perturbations covering the whole tree demonstrate a model based on random synchronisation of the evaluation of the parent and child which cause parent and offspring evaluations to be identical. This predicts the effectiveness of fitness measurement grows slowly as O(log(n)) with number n of test cases. This geometric distribution model is tested on genetic programming symbolic regression.

Type: Proceedings paper
Title: Measuring failed disruption propagation in genetic programming
Event: GECCO '23: Genetic and Evolutionary Computation Conference
Dates: 9 Jul 2022 - 13 Jul 2022
ISBN-13: 9781450392372
Open access status: An open access version is available from UCL Discovery
DOI: 10.1145/3512290.3528738
Publisher version: https://doi.org/10.1145/3512290.3528738
Language: English
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords: mutational robustness, antifragile correctness attraction, SBSE, software resilience, information theory, entropy
UCL classification: UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL
URI: https://discovery.ucl.ac.uk/id/eprint/10154055
Downloads since deposit
53Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item