Evolving a CUDA Kernel from an nVidia Template

Created by W.Langdon from gp-bibliography.bib Revision:1.4504

  author =       "W. B. Langdon and M. Harman",
  title =        "Evolving a CUDA Kernel from an nVidia Template",
  institution =  "Department of Computer Science, Crest Centre, King's
                 College, London",
  year =         "2010",
  number =       "TR-10-07",
  address =      "Strand, London, WC2R 2LS, UK",
  month =        "6 " # jul,
  note =         "Invited presentation to the Natural Computing
                 Applications Forum, http://www.ncaf.org.uk, 12-13 July
                 2010, Surrey University",
  keywords =     "genetic algorithms, genetic programming, gipsy,
                 strongly typed genetic programming, grammar, SBSE,
                 software engineering, GPU",
  URL =          "http://www.dcs.kcl.ac.uk/technical-reports/papers/TR-10-07.pdf",
  abstract =     "We automatically generate an nVidia parallel CUDA
                 graphics card kernel with identical functionality to
                 existing highly optimised ancient sequential C code.
                 Essentially generic GPGPU C++ code supplied by the
                 hardware manufacturer is converted into a BNF grammar.
                 Strongly typed genetic programming uses the BNF to
                 generate compilable and runnable graphics card kernels,
                 which terminate. Their fitness is given by running the
                 population on a GPU against a test suite used to test
                 the original sequential code.",
  size =         "2 pages",

Genetic Programming entries for William B Langdon Mark Harman