- author = "Ann Thorhauer and Franz Rothlauf",
- title = "On the Bias of Syntactic Geometric Recombination in Genetic Programming and Grammatical Evolution",
- booktitle = "GECCO '15: Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation",
- year = "2015",
- editor = "Sara Silva and Anna I Esparcia-Alcazar and Manuel Lopez-Ibanez and Sanaz Mostaghim and Jon Timmis and Christine Zarges and Luis Correia and Terence Soule and Mario Giacobini and Ryan Urbanowicz and Youhei Akimoto and Tobias Glasmachers and Francisco {Fernandez de Vega} and Amy Hoover and Pedro Larranaga and Marta Soto and Carlos Cotta and Francisco B. Pereira and Julia Handl and Jan Koutnik and Antonio Gaspar-Cunha and Heike Trautmann and Jean-Baptiste Mouret and Sebastian Risi and Ernesto Costa and Oliver Schuetze and Krzysztof Krawiec and Alberto Moraglio and Julian F. Miller and Pawel Widera and Stefano Cagnoni and JJ Merelo and Emma Hart and Leonardo Trujillo and Marouane Kessentini and Gabriela Ochoa and Francisco Chicano and Carola Doerr",
- isbn13 = "978-1-4503-3472-3",
- pages = "1103--1110",
- keywords = "genetic algorithms, genetic programming, grammatical evolution",
- month = "11-15 " # jul,
- organisation = "SIGEVO",
- address = "Madrid, Spain",
-
URL = "
http://doi.acm.org/10.1145/2739480.2754726",
-
DOI = "
doi:10.1145/2739480.2754726",
- publisher = "ACM",
- publisher_address = "New York, NY, USA",
-
abstract = "For fixed-length binary representations as used in
genetic algorithms, standard recombination operators
(e.g.,~one-point crossover) are unbiased. Thus, the
application of recombination only reshuffles the
alleles and does not change the statistical properties
in the population. Using a geometric view on
recombination operators, most search operators for
fixed-length strings are geometric, which means that
the distances between offspring and their parents are
less than, or equal to, the distance between their
parents. In genetic programming (GP) and grammatical
evolution (GE), the situation is different since the
recombination operators are applied to variable-length
structures. Thus, most recombination operators for GE
and GP are not geometric.
This paper focuses on the bias of recombination in GE and GP and studies whether the application of recombination alone produces specific types of solutions with a higher probability. We consider two different types of recombination operators: standard recombination and syntactic geometric recombination. In our experiments, we performed random walks through the binary tree search space and found that syntactic geometric recombination operators are biased and strongly reduce population diversity. In a performance comparison, we found that syntactic geometric recombination leads to large fitness improvements in the first generations, but that fitness converges after several generations and no further search is possible.",
- notes = "Also known as \cite{2754726} GECCO-2015 A joint meeting of the twenty fourth international conference on genetic algorithms (ICGA-2015) and the twentith annual genetic programming conference (GP-2015)",
