In genetic programming (GP), programs are usually evaluated by applying them to tests, and fitness function indicates only how many of them have been passed. We posit that scrutinising the outcomes of programs interactions with individual tests may help making program synthesis more effective. To this aim, we propose DOC, a method that autonomously derives new search objectives by clustering the outcomes of interactions between programs in the population and the tests. The derived objectives are subsequently used to drive the selection process in a single or multiobjective fashion. An extensive experimental assessment on 15 discrete program synthesis tasks representing two domains shows that DOC significantly outperforms conventional GP and implicit fitness sharing.
@INPROCEEDINGS { Krawiec:2015:EuroGP,
ABSTRACT = { In genetic programming (GP), programs are usually evaluated by applying them to tests, and fitness function indicates only how many of them have been passed. We posit that scrutinising the outcomes of programs interactions with individual tests may help making program synthesis more effective. To this aim, we propose DOC, a method that autonomously derives new search objectives by clustering the outcomes of interactions between programs in the population and the tests. The derived objectives are subsequently used to drive the selection process in a single or multiobjective fashion. An extensive experimental assessment on 15 discrete program synthesis tasks representing two domains shows that DOC significantly outperforms conventional GP and implicit fitness sharing. },
ADDRESS = { Copenhagen },
AUTHOR = { Krzysztof Krawiec and Pawel Liskowski },
BOOKTITLE = { 18th European Conference on Genetic Programming },
DOI = { doi:10.1007/978-3-319-16501-1_5 },
EDITOR = { Penousal Machado and Malcolm I. Heywood and James McDermott and Mauro Castelli and Pablo Garcia-Sanchez and Paolo Burelli and Sebastian Risi and Kevin Sim },
ISBN13 = { 978-3-319-16500-4 },
KEYWORDS = { genetic algorithms, genetic programming, Program synthesis, Test-based problems, Multiobjective evolutionary computation },
MONTH = { {8-10 } # apr },
NOTES = { Part of \cite{Machado:2015:GP} EuroGP'2015 held in conjunction with EvoCOP2015, EvoMusArt2015 and EvoApplications2015 },
ORGANISATION = { EvoStar },
PAGES = { 53--65 },
PUBLISHER = { Springer },
SERIES = { LNCS },
TITLE = { Automatic Derivation of Search Objectives for Test-Based Genetic Programming },
VOLUME = { 9025 },
YEAR = { 2015 },
1 = { https://doi.org/10.1007/978-3-319-16501-1_5 },
}