A linear estimation-of-distribution GP system

Riccardo Poli, Nicholas Freitag McPhee

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    18 Scopus citations


    We present N-gram GP, an estimation of distribution algorithm for the evolution of linear computer programs. The algorithm learns and samples a joint probability distribution of triplets of instructions (or 3-grams) at the same time as it is learning and sampling a program length distribution. We have tested N-gram GP on symbolic regressions problems where the target function is a polynomial of up to degree 12 and lawn-mower problems with lawn sizes of up to 12×12. Results show that the algorithm is effective and scales better on these problems than either linear GP or simple stochastic hill-climbing.

    Original languageEnglish (US)
    Title of host publicationGenetic Programming - 11th European Conference, EuroGP 2008, Proceedings
    Number of pages12
    StatePublished - Jul 21 2008
    Event11th European Conference on Genetic Programming, EuroGP 2008 - Naples, Italy
    Duration: Mar 26 2008Mar 28 2008

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume4971 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349


    Conference11th European Conference on Genetic Programming, EuroGP 2008


    Dive into the research topics of 'A linear estimation-of-distribution GP system'. Together they form a unique fingerprint.

    Cite this