An Evolutionary Metagraph Approach for Solving Problems in Complex Subject Areas

• Published in: International Youth Conference on Radio Electronics, Electrical and Power Engineering (REEPE ...) (Online) pp. 1 - 6
• Main Authors: Nardid, Anatoly N., Vinnikov, Stepan S., Orazov, Alexey V., Gapanyuk, Yuriy E.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.04.2025
• Subjects: Benchmark testing; cartesian genetic programming (CGP); Complexity theory; Directed acyclic graph; directed acyclic graphs (DAGs); evolutionary algorithms; Evolutionary computation; Genetic programming; linear genetic programming (LGP); metagraphs; Power engineering; Reproducibility of results
• ISSN: 2831-7262
• DOI: 10.1109/REEPE63962.2025.10970829

This paper explores the integration of metagraphs with genetic programming (GP), offering a novel perspective aimed at overcoming the limitations of traditional graph-based approaches. Metagraphs, providing an advanced abstraction over standard graphs, are examined for their potential to more efficiently encapsulate complex relationships and hierarchical structures. The focus is on how metagraphs can contribute to evolutionary algorithms by enriching the representation of problem spaces, potentially leading to improved adaptability and precision in solutions. We discuss the initial theoretical insights and potential benefits of this integration, positioning metagraphs as a promising tool for enhancing the effectiveness of evolutionary strategies. This exploration is intended to pave new research pathways in GP, proposing that metagraphs hold the potential to significantly augment the outcomes of evolutionary processes.