Summary
We describe a framework for program evolution with an EDA-based approach. In this framework, the probability distribution of programs is estimated using a Bayesian network, and individuals are generated based on the estimated distribution. Considering that a dependency relationship of nodes in a program tree is explicit, i.e. the dependency relationship is strong between a parent node and its child node in a program expressed as a tree structure, we have chosen a Bayesian network as the distribution model of programs.
In order to demonstrate the effectiveness of our approach, this chapter shows results of comparative experiments with Genetic Programming. Thereafter, we discuss how Estimation of Distribution Programming works and the transitions of the evolved programs that are the forte of our methods. We also analyze the performance of a hybrid system which combines Estimation of Distribution Programming and Genetic Programming.
Access provided by Autonomous University of Puebla. Download to read the full chapter text
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
P. J. Angeline. Two selfadaptive crossover operations for genetic programming. In Advances in Genetic Programming II. MIT Press, 1995.
P. J. Angeline. Subtree crossover causes bloat. In Genetic Programming 1998: Proceedings of the Third Annual Conference, pp. 745–752. Morgan Kaufmann, 22–25 July 1998.
S. Baluja. Population-based incremental learning: A method for integrating genetic search based function optimization and competitive learning. Technical Report CMU-CS-94-163, Carnegie Mellon University, Pittsburgh, PA, 1994.
B. Cestnik. Estimating probabilities: A crucial task in machine learning. In Proceedings of the 9th European Conference on Artificial Intelligence, pp. 147–149, 1990.
J. M. Daida, H. Li, R. Tang, and A. M. Hilss. What makes a problem GP-hard? validating a hypothesis of structural causes. In Proceedings of the Genetic and Evolutionary Computation Conference GECCO-2004, pp. 1665–1677. Springer-Verlag, 2003.
H. Iba and H. de Garis. Extending genetic programming with recombinative guidance. In Advances in Genetic Programming 2, pp. 69–88. MIT Press, 1995.
C. Igel and K. Chellapilla. Investigating the influence of depth and degree of genotypic change on fitness in genetic programming. In Proceedings of the Genetic and Evolutionary Computation Conference GECCO-1999, volume 2, pp. 1061–1068. Morgan Kaufmann, 13–17 July 1999.
T. Ito, H. Iba, and S. Sato. Depth-dependent crossover for genetic programming. In Proceedings of the 1998 IEEE World Congress on Computational Intelligence, pp. 775–780. IEEE Press, 5–9 May 1998.
J. R. Koza. Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, 1992.
W. B. Langdon. Size fair and homologous tree genetic programming crossovers. In Proceedings of the Genetic and Evolutionary Computation Conference GECCO-1999, volume 2, pp. 1092–1097. Morgan Kaufmann, 13–17 1999.
W. B. Langdon and R. Poli. Genetic programming bloat with dynamic fitness. In Proceedings of the First European Workshop on Genetic Programming. Springer-Verlag, 1998.
P. Larrñnaga, R. Etxeberria, J. A. Lozano, and J. M. Peña. Combinatorial optimization by learning and simulation of Bayesian networks. In Proceedings of the Conference in Uncertainty in Artificial Intelligence: UAI-2000, pp. 343–352, 2000.
P. Larrañaga, R. Etxeberria, J. A. Lozano, and J.M. Peña. Optimization in continuous domains by learning and simulation of Gaussian networks. In Proceedings of the Workshop in Optimization by Building and Using Probabilistic Models, pp. 201–204, 2000.
T. K. Paul and H. Iba. Reinforcement learning estimation of distribution algorithm. In Proceedings of Genetic and Evolutionary Computation Conference GECCO-2003. Springer-Verlag, 2003.
T. K. Paul and H. Iba. Selection of the most useful subset of genes for gene expression-based classification. In Proceedings of Congress on Evolutionary Computation: CEC-2004, 2004.
R. Poli and W. B. Langdon. On the search properties of different crossover operators in genetic programming. In Genetic Programming 1998: Proceedings of the Third Annual Conference, pp. 293–301. Morgan Kaufmann, 22–25 1998.
R. Poli and W. B. Langdon. Foundations of Genetic Programming. Springer-Verlag, 2002.
W. F. Punch, D. Zongker, and E. D. Goodman. The royal tree problem, a benchmark for single and multiple population genetic programming. In Advances in Genetic Programming II, pp.299–316. MIT Press, 1995.
R. P. Salustowicz and J. Schmidhuber. Probabilistic incremental program evolution: Stochastic search through program space. In M. van Someren and G. Widmer, editors, Machine Learning: ECML-97, volume 1224, pp. 213–220. Springer-Verlag, 1997.
T. Yabuki and H. Iba. Genetic programming using a Turing complete representation: recurrent network consisting of trees. In L. Nunes de Castro and F. J. Von Zuben, editors, Recent Developments in Biologically Inspired Computing. Idea Group Inc., 2004. (to be published).
K. Yanai and H. Iba. Estimation of distribution programming based on Bayesian networks. In Proceedings of Congress on Evolutionary Computation: CEC-2003, pp. 1618–1625, 2003.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Yanai, K., Iba, H. (2006). Estimation of Distribution Programming: EDA-based Approach to Program Generation. In: Lozano, J.A., Larrañaga, P., Inza, I., Bengoetxea, E. (eds) Towards a New Evolutionary Computation. Studies in Fuzziness and Soft Computing, vol 192. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-32494-1_5
Download citation
DOI: https://doi.org/10.1007/3-540-32494-1_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-29006-3
Online ISBN: 978-3-540-32494-2
eBook Packages: EngineeringEngineering (R0)