Abstract
We have proposed a novel Estimation of Distribution Algorithms with graph kernels (EDA-GK). By using the graph kernels, we can search for solutions in a feature space. The use of the graph kernel can eliminate the effect of the ruggedness of genotype-phenotype mappings of evolutionary algorithms. In this paper, we extend the EDA-GK to cope with graphs with node types. In order to achieve this, the histogram in the kernel density estimation function is separated into several sub-histograms for modes of inter-types, and for nodes of each type. Experimental results on the Edge-Max problems show the effectiveness of the proposed method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
P. Larrañaga, and J.A. Lozano, Estimation of Distribution Algorithms, Kluwer Academic Publishers, 2003
H. Handa. Use of graph kernels in Estimation of Distribution Algorithms. In Proc. of 2012 IEEE Congress on Evolutionary Computation (CEC), pages 1–6, 2012.
E. Alba, and F. Chicano, ACOhg: Dealing with huge graphs, Proc. the 2007 ACM Genetic and Evolutionary Conference, pp. 10–17, 2007.
F. Chicano, and E. Alba, Ant colony optimization with partial order reduction for discovering safety property violations in concurrent models, Information Processing Letters, Vol. 106, No. 6, pp. 221–231, 2008.
J. McDermott,U.-M. O’Reilly, An executable graph representation for evolutionary generative music, Proc. the 2011 ACM Genetic and Evolutionary Conference, pp. 403–412, 2011.
T.E. Lewis, and G.D. Magoulas, Strategies to minimise the total run time of cyclic graph based genetic programming with GPUs, Proc. the 2009 ACM Genetic and Evolutionary Conference, pp. 1379–1386, 2009.
S. Shirakawa, and T. Nagao, Graph structured program evolution with automatically defined nodes, Proc. the 2009 ACM Genetic and Evolutionary Conference, pp. 1107–1115, 2009.
S. Mabu, K. Hirasawa, and J.Hu, A Graph-Based Evolutionary Algorithm: Genetic Network Programming (GNP) and Its Extension Using Reinforcement Learning, Evolutionary Computation, Vol. 15, No. 3, pp. 369–398, 2007.
H. Kashima, K. Tsuda, and A. Inokuchi, Marginalized Kernels Between Labeled Graphs, Proc. 20th International Conference on Machine Learning, pp. 321–328, 2003.
K.M. Borgwardt,and H.-P. Kriegel, Shortest-path kernels on graphs, Proc. 5th International Conference Data Mining, 2005.
Acknowledgments
This work was partially supported by the Grant-in-Aid for Young Scientists (B) and the Grant-in-Aid for Scientific Research (C) of MEXT, Japan (23700267, 26330291).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Maezawa, K., Handa, H. (2017). Estimation of Distribution Algorithms with Graph Kernels for Graphs with Node Types. In: Leu, G., Singh, H., Elsayed, S. (eds) Intelligent and Evolutionary Systems. Proceedings in Adaptation, Learning and Optimization, vol 8. Springer, Cham. https://doi.org/10.1007/978-3-319-49049-6_18
Download citation
DOI: https://doi.org/10.1007/978-3-319-49049-6_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-49048-9
Online ISBN: 978-3-319-49049-6
eBook Packages: EngineeringEngineering (R0)