A Compiling Genetic Programming System that Directly Manipulates the Machine Code

Created by W.Langdon from gp-bibliography.bib Revision:1.4524

  author =       "Peter Nordin",
  title =        "A Compiling Genetic Programming System that Directly
                 Manipulates the Machine Code",
  booktitle =    "Advances in Genetic Programming",
  publisher =    "MIT Press",
  editor =       "Kenneth E. {Kinnear, Jr.}",
  year =         "1994",
  pages =        "311--331",
  chapter =      "14",
  URL =          "http://www.amazon.co.uk/Advances-Genetic-Programming-Complex-Adaptive/dp/0262111888",
  URL =          "http://cognet.mit.edu/sites/default/files/books/9780262277181/pdfs/9780262277181_chap14.pdf",
  size =         "19 pages",
  keywords =     "genetic algorithms, genetic programming",
  abstract =     "Most genetic programming approaches use a technique
                 where a problem specific language is executed by an
                 interpreter. The individual code segments in the
                 population are decoded at run time by a virtual
                 machine. The disadvantage of this paradigm is that
                 interpreting the program involves a large overhead.
                 Often the complete system and the genetic operators
                 themselves are written in an interpreting language like
                 LISP. This reduces performance in most hardware
                 environments. We have evaluated the idea of using the
                 lowest level native binary machine code as the programs
                 in the population. There is no intermediate language or
                 any interpreting steps. The genetic program that
                 administers these machine code segments is written in
                 the C-language. The algorithm is of steady state type
                 and uses a small tournament as the selection mechanism.
                 This approach has enhanced performance by a magnitude
                 of three compared to a conventional system in an
                 interpreting language. The increased performance is
                 tested on a problem of symbolic regression of a
                 classifier function in machine code. We evolved a
                 machine code program that classifies Swedish words into
                 nouns and non-nouns by spelling only. We compare the
                 compiling genetic programming system (COPS) with a
                 Neural Network performing the same task. In our
                 example, the results show superior performance of the
                 COPS compared to the connectionist approach. While the
                 classification and generalisation capabilities were
                 equal, the training time was more than 200 times
                 faster, the classification time 500 times faster and
                 the memory requirements are at least 10 times lower
                 with the COPS, as compared with the Neural Network.",
  notes =        "Machine code GP Sun Spark and i868",

Genetic Programming entries for Peter Nordin