Abstract
Gaussian process (GP) has fewer parameters, simple model and output of probabilistic sense, when compared with the methods such as support vector machines. Selection of the hyper-parameters is critical to the performance of Gaussian process model. However, the common-used algorithm has the disadvantages of difficult determination of iteration steps, over-dependence of optimization effect on initial values, and easily falling into local optimum. To solve this problem, a method combining the Gaussian process with memetic algorithm was proposed. Based on this method, memetic algorithm was used to search the optimal hyper parameters of Gaussian process regression (GPR) model in the training process and form MA-GPR algorithms, and then the model was used to predict and test the results. When used in the marine long-range precision strike system (LPSS) battle effectiveness evaluation, the proposed MA-GPR model significantly improved the prediction accuracy, compared with the conjugate gradient method and the genetic algorithm optimization process.
Similar content being viewed by others
References
SUN Bin, YAO Hai-tao, LIU Ting. Prediction of short-term wind speed based on Gaussian process [J]. Proceedings of the CSEE, 2012, 32(29): 104–109. (in Chinese)
SUN Zong-hai, SHEN Qian-qian. Study of the online modeling problem based on Gaussian process [D]. Guangzhou: South China University of Technology, 2011. (in Chinese)
SU Guo-shao, QUAN Jiang. A cooperative optimization algorithm based on gaussi an process and particle swarm optimization for optimizing expensive problems [C]// International Joint Conference on Computational Sciences and Optimization. Los Alamitos: IEEE Computer Society, 2009: 929–933.
SU Guo-shao, ZHANG Ke-shi, LU Hai-bo. Particle swarm optimization in displacement back analysis-coordinated optimization technique of Gaussian process [J]. Rock & Soil Mechanics, 2011, 32(2): 510–515. (in Chinese)
XU Chong, LIU Bao-guo, LIU Kai-yun, GUO Jia-qi. Intelligent model for analysis & prediction of landslide displacement time series based on particle swarm-Gaussian process regression coupling algorithm [J]. Rock & Soil Mechanics, 2011, 32(6): 1669–1675. (in Chinese)
LIU Kai-yun, FANG Yu, LIU Bao-guo, XU Chong. Genetic Gaussian process regression model for prediction of tunnel surrounding rock deformation [J]. Journal of the China Railway Society, 2011, 33(12): 101–106. (in Chinese)
XU Chong, LIU Bao-guo, LIU Kai-yun, GUO Jia-qi. Intelligent design of Gaussian process slope angle based on compound kernel function [J]. Rock & Soil Mechanics, 2010, 31(3): 821–826. (in Chinese)
TENNE Y, STEVEN W A F. A framework for memetic optimization using variable global and local surrogate models [J]. Soft Computing, 2009, 13(8/9): 781–793.
PADMAVATHI S, MOHITGOLCHHA S, SEENIMOHAMED A. Memetic algorithm based task scheduling using probabilistic local search [C]// Swarm, Evolutionary, and Memetic Computing. Berlin Heidelberg: Springer, 2012: 224–231.
MOSCATO P, CARLOS C. A gentle introduction to memetic algorithms [M]. New York: US, Springer, 2003: 105–144.
BERGSTRA J, YOSHUA B. Random search for hyper-parameter optimization [J]. The Journal of Machine Learning Research, 2012, 13: 281–305.
RASMUSSEN C E, CHRISTOPHER K I. Gaussian processes for machine learning [M]. Boston: America: The MIT Press, 2005.
YAO Fu-tian. Study of hyper-spectral image classification based on gaussian process [D]. Hangzhou: Zhejiang University, 2011. (in Chinese)
MOSCATO P. On evolution, search, optimization, genetic algorithms and martial arts: Towards memetic algorithms [R]. Caltech Concurrent Computation Program, C 3P Report 826, 1989.
MELANIE M. An introduction to genetic algorithms [M]. Cambridge, England: The MIT Press, 1999.
LI Yun, LIU Gang, LAO Song-yang. A genetic algorithm for community detection in complex networks [J]. Journal of Central South University, 2013, 20(5): 1269–1276.
ZHAO Wan-zhong, WANG Chun-yan, YU Lei-yan, CHEN Tao. Performance optimization of electric power steering based on multi-objective genetic algorithm [J]. Journal of Central South University, 2013, 20(1): 98–104.
RODRIGUEZ T E, JOSE T J. Memetic algorithms for constructing binary covering arrays of strength three [C]// 9th International Conference on Artificial Evolution. Berlin Heidelberg: Springer, 2010: 86–97.
HAUPT R L, WERNER D H. Genetic algorithms in electro magnetics [M]. New Jersey: John Willy & Sons, 2007.
GUIMARAES F G, CAMPELO F, IGARASHI H, LOWTHER D A, RAMIREZ J A. Optimization of cost functions using evolutionary algorithms with local learning and local search [J]. Magnetics, IEEE Transactions on, 2007, 43(4): 1641–1644.
MICHALEWICA Z. Genetic algorithms data structures evolution programs [M]. Berlin: Springer-Verlag, 1996.
Author information
Authors and Affiliations
Corresponding author
Additional information
Foundation item: Project(513300303) supported by the General Armament Department, China
Rights and permissions
About this article
Cite this article
Zhang, L., Liu, Z., Zhang, Jq. et al. A genetic Gaussian process regression model based on memetic algorithm. J. Cent. South Univ. 20, 3085–3093 (2013). https://doi.org/10.1007/s11771-013-1832-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11771-013-1832-0