Graph representations in genetic programming

Author(s)

Sotto, Leo Francoso Dal Piccol ^[1] ; Kaufmann, Paul ^[2] ; Atkinson, Timothy ^[3] ; Kalkreuth, Roman ^[4] ; Basgalupp, Marcio Porto ^[5]

Affiliation

^[1] Fraunhofer SCAI, St Augustin - Germany

^[2] Johannes Gutenberg Univ Mainz, Mainz - Germany

^[3] NNAISENSE SA, Lugano - Switzerland

^[4] TU Dortmund, Dortmund - Germany

^[5] Univ Fed Sao Paulo, Sao Jose Dos Campos, SP - Brazil

• Document type: Journal article
• Source: Genetic Programming and Evolvable Machines; v. 22, n. 4, SI SEP 2021.
• Web of Science Citations: 0

Abstract

Graph representations promise several desirable properties for genetic programming (GP); multiple-output programs, natural representations of code reuse and, in many cases, an innate mechanism for neutral drift. Each graph GP technique provides a program representation, genetic operators and overarching evolutionary algorithm. This makes it difficult to identify the individual causes of empirical differences, both between these methods and in comparison to traditional GP. In this work, we empirically study the behaviour of Cartesian genetic programming (CGP), linear genetic programming (LGP), evolving graphs by graph programming and traditional GE. By fixing some aspects of the configurations, we study the performance of each graph GP method and GP in combination with three different EAs: generational, steady-state and (1 + lambda). In general, we find that the best choice of representation, genetic operator and evolutionary algorithm depends on the problem domain. Further, we find that graph GP methods can increase search performance on complex real-world regression problems and, particularly in combination with the (1 + lambda) EA, are significantly better on digital circuit synthesis tasks. We further show that the reuse of intermediate results by tuning LGP's number of registers and CGP's levels back parameter is of utmost importance and contributes significantly to better convergence of an optimization algorithm when solving complex problems that benefit from code reuse. (AU)

• FAPESP's process: 16/07095-5 - Development of the probabilistic linear genetic programming technique and application on Kaizen programming for supervised machine learning
• Grantee: Léo Françoso Dal Piccol Sotto
• Support Opportunities: Scholarships in Brazil - Doctorate (Direct)