Mixed IFS: Resolution of the Inverse Problem Using Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4549

  author =       "Evelyne Lutton and Jacques Levy-Vehel and 
                 Guillaume Cretin and Philippe Glevarec and Cedric Roll",
  title =        "Mixed {IFS}: Resolution of the Inverse Problem Using
                 Genetic Programming",
  institution =  "Inria",
  year =         "1995",
  type =         "Research Report",
  number =       "No 2631",
  keywords =     "genetic algorithms, genetic programming",
  URL =          "http://hal.inria.fr/inria-00074056/en/",
  URL =          "http://hal.inria.fr/docs/00/07/40/56/PDF/RR-2631.pdf",
  URL =          "http://citeseer.ist.psu.edu/cretin95mixed.html",
  URL =          "http://citeseerx.ist.psu.edu/viewdoc/summary?doi=",
  abstract =     "We address here the resolution of the so-called
                 inverse problem for IFS. This problem has already been
                 widely considered, and some studies have been performed
                 for affine IFS, using deterministic or stochastic
                 methods (Simulated Annealing or Genetic Algorithm).
                 When dealing with non affine IFS, the usual techniques
                 do not perform well, except if some  a priori
                 hypotheses on the structure of the IFS (number and type
                 functions) are made. In this work, a Genetic
                 Programming method is investigated to solve the
                 "general" inverse problem, which permits to
                 perform at the same time a numeric and a symbolic
                 optimization. The use of "mixed IFS", as we
                 call them, may enlarge the scope of some applications,
                 as for example image compression, because they allow to
                 code a wider range of shapes.",
  notes =        "Mainly in english, abstract also en francaise Use
                 distance masks for deciding how close GP is to target
                 image (part of fitness function). Says {"}The distance
                 images are very efficient{"} [page 12]. Mutation of
                 constants by +/-10% and variables to constants. Notes
                 constants {"}disappear{"} from the population. popsize
                 20 to 50 and 1000 to 2000 generations [page 10]. GP
                 functions {"}does not resemble the one [used to create]
                 the target images{"} [page 12]. {"}GP algorith, which
                 seems to perform a more efficient search in a large
                 space.{"} [page 16]",
  size =         "17 pages. See also \cite{lutton:1995:IFScs} and

Genetic Programming entries for Evelyne Lutton Jacques Levy-Vehel Guillaume Cretin Philippe Glevarec Cedric Roll