An improved $\lambda$-linear genetic programming evaluated in solving the Santa Fe ant trail problem

Created by W.Langdon from gp-bibliography.bib Revision:1.4340

  author =       "Leo Francoso Dal Piccol Sotto and 
                 Vinicius Veloso {de Melo} and Marcio P. Basgalupp",
  title =        "An improved {$\lambda$}-linear genetic programming
                 evaluated in solving the Santa Fe ant trail problem",
  booktitle =    "Proceedings of the 31st Annual {ACM} Symposium on
                 Applied Computing",
  publisher =    "ACM",
  year =         "2016",
  editor =       "Sascha Ossowski",
  address =      "Pisa, Italy",
  month =        apr # " 4-8",
  pages =        "103--108",
  isbn13 =       "978-1-4503-3739-7",
  keywords =     "genetic algorithms, genetic programming, linear
                 genetic programming, santa fe ant trail",
  bibdate =      "2016-06-07",
  bibsource =    "DBLP,
  URL =          "",
  DOI =          "doi:10.1145/2851613.2851669",
  acmid =        "2851669",
  size =         "6 pages",
  abstract =     "We propose in this paper a new approach called
                 lambda-LGP (lambda-Linear Genetic Programming), a
                 variation of the well-know LGP (Linear Genetic
                 Programming) algorithm. Starting with an LGP based only
                 on effective macro and micromutations, the l-LGP
                 proposed in this work consists in extending the way in
                 which the individuals are chosen for reproduction. In
                 this model, a constant number (l) of a particular kind
                 of mutation is applied to each individual, thus
                 exploring its neighbouring fitness regions, and might
                 be replaced by one of its children according to
                 different criteria. Several configurations were tested
                 in the benchmark problem known as Santa Fe Ant Trail.
                 Results obtained show a very significant improvement by
                 using this proposed variation. For the Ant Trail
                 problem, lambda-LGP outperformed not only LGP, but also
                 several state-of-the-art methods.",

Genetic Programming entries for Leo Francoso Dal Piccol Sotto Vinicius Veloso de Melo Marcio Porto Basgalupp