Krzysztof Krawiec


Home

Research:

edit SideBar

The instructions used for solving typical genetic programming tasks have strong mathematical properties. In this study, we leverage one of such properties: invertibility. A search operator is proposed that performs an approximate reverse execution of program fragments, trying to determine in this way the desired semantics (partial outcome) at intermediate stages of program execution. The desired semantics determined in this way guides the choice of a subprogram that replaces the old program fragment. An extensive computational experiment on 20 symbolic regression and Boolean domain problems leads to statistically significant evidence that the proposed Random Desired Operator outperforms all typical combinations of conventional mutation and crossover operators.

@INPROCEEDINGS { Wieloch:2013:RPB:2463372.2463493,
    AUTHOR = { Wieloch, Bartosz and Krawiec, Krzysztof },
    TITLE = { Running programs backwards: instruction inversion for effective search in semantic spaces },
    BOOKTITLE = { Proceeding of the fifteenth annual conference on Genetic and evolutionary computation conference },
    SERIES = { GECCO '13 },
    YEAR = { 2013 },
    ISBN = { 978-1-4503-1963-8 },
    LOCATION = { Amsterdam, The Netherlands },
    PAGES = { 1013--1020 },
    NUMPAGES = { 8 },
    URL = { http://doi.acm.org/10.1145/2463372.2463493 },
    DOI = { 10.1145/2463372.2463493 },
    ACMID = { 2463493 },
    PUBLISHER = { ACM },
    ADDRESS = { New York, NY, USA },
    KEYWORDS = { desired semantics, genetic programming, instruction inversion, program semantics, search operators },
    ABSTRACT = { The instructions used for solving typical genetic programming tasks have strong mathematical properties. In this study, we leverage one of such properties: invertibility. A search operator is proposed that performs an approximate reverse execution of program fragments, trying to determine in this way the desired semantics (partial outcome) at intermediate stages of program execution. The desired semantics determined in this way guides the choice of a subprogram that replaces the old program fragment. An extensive computational experiment on 20 symbolic regression and Boolean domain problems leads to statistically significant evidence that the proposed Random Desired Operator outperforms all typical combinations of conventional mutation and crossover operators. },
}


Powered by PmWiki