Vanishing Ideal Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4420

  author =       "Hiroshi Kera and Hitoshi Iba",
  title =        "Vanishing Ideal Genetic Programming",
  booktitle =    "Proceedings of 2016 IEEE Congress on Evolutionary
                 Computation (CEC 2016)",
  year =         "2016",
  editor =       "Yew-Soon Ong",
  pages =        "5018--5025",
  address =      "Vancouver",
  month =        "24-29 " # jul,
  publisher =    "IEEE Press",
  keywords =     "genetic algorithms, genetic programming, VIGP",
  isbn13 =       "978-1-5090-0623-6",
  DOI =          "doi:10.1109/CEC.2016.7748325",
  size =         "8 pages",
  abstract =     "In symbolic regression, which aims to find a function
                 that satisfies the target values for all data points,
                 one of the major challenges is that the solutions
                 cannot be uniquely determined. Genetic programming (GP)
                 provides a powerful approach to symbolic regression in
                 that it does not require models of functions to be
                 fixed. However, it is known that GP suffers from a
                 phenomenon known as bloat, meaning that candidate
                 functions attain an excessively complicated form during
                 the search, which is undesirable in many applications.
                 While the majority of approaches for regulating bloat
                 introduce anti-bloat genetic operators or anti-bloat
                 selection schemes, most of these are derived from
                 heuristics and/or require well-tuned hyperparameters.
                 In the present study, we propose a novel approach in
                 which genetic trees of GP are reduced during the search
                 using a basis of a set of polynomials (vanishing ideal)
                 that are equivalent to zero for the data points of
                 symbolic regression. The vanishing ideal is computed
                 using an algebraic approach, and because it only
                 requires data points as input, our approach does not
                 involve the tuning of any hyper-parameters. The
                 proposed approach regulates bloat and efficiently
                 determines simple solutions. We compare our approach
                 with standard GP with a penalty term for the height of
                 trees in the fitness, and demonstrate the effectiveness
                 of our approach to two tasks (real-valued symbolic
                 regression and the 6-parity problem).",
  notes =        "'fitness is based only on .. [at] the data points'
                 'Removal of vanishing polynomials might be similar to
                 removal of introns'

                 Genetic repair? Unclear how vanishing polynomial is
                 removed from GP tree. Buchberger-Moeller (BM)
                 algorithm. Taylor v. Pade. Only rational polynomials.
                 Mostly continuous eg Keijzer-15 but also looks at
                 6-parity (with XOR). DEAP


Genetic Programming entries for Hiroshi Kera Hitoshi Iba