Prediction of Pareto Dominance Using Nearest Neighbor Method Based on Decision Space Transformation

被引：0
|
作者：
Li W.-B. [1 ,2 ]
He J.-J. [1 ]
Feng C.-Y. [2 ]
Guo G.-Q. [2 ]
机构：
[1] College of Information Science and Engineering, Central South University, Changsha
[2] College of Information and Communication Engineering, Hunan Institute of Science and Technology, Yueyang
来源：
Guo, Guan-Qi (gq.guo@163.com) | 1600年 / Science Press卷 / 43期
基金：
中国国家自然科学基金;
关键词：
Nearest neighbor method; Pareto dominance; Space transformation; Tendency model;
D O I：
10.16383/j.aas.2017.c150877
中图分类号：
学科分类号：
摘要：
In this paper, nearest neighbor prediction is used to decide Pareto dominance of candidate solutions in expensive multi-objective optimization. For improving the accuracy of predicting Pareto dominance in decision space, a transformation method of decision space is proposed. Based on correlation analysis of objective functions and decision attributes, computationally efficient attribute tendency models of objective functions are set up. These models are used to re-construct the decision space such that the knowledge of objective space is introduced and the neighborhood relation in decision space can more effectively reflect that in the original objective space. Experiments of Pareto dominance prediction are carried out on a group of typical multi-objective optimization problems. The results indicate that the prediction of Pareto dominance using the proposed method is more accurate and efficient than the existing methods. Copyright © 2017 Acta Automatica Sinica. All rights reserved.
引用
收藏
页码：294 / 301
页数：7
相关论文
共 19 条
  • [1] Gong M.-G., Jiao L.-C., Yang D.-D., Ma W.-P., Research on evolutionary multi-objective optimization algorithms, Journal of Software, 20, 2, pp. 271-289, (2009)
  • [2] Zhan W., Research on Manifold Learning Algorithm for High-Dimensional Multi-objective Optimization Problems, (2013)
  • [3] Tesch M., Schneider J., Choset H., Expensive multiobjective optimization for robotics, Proceedings of the 2013 IEEE International Conference on Robotics and Automation, pp. 973-980, (2013)
  • [4] Douguet D., e-LEA3D: a computational-aided drug design web server, Nucleic Acids Research, 38, pp. W615-W621, (2010)
  • [5] Zhang Q.F., Liu W.D., Tsang E., Virginas B., Expensive multiobjective optimization by MOEA/D with Gaussian process model, IEEE Transactions on Evolutionary Computation, 14, 3, pp. 456-474, (2010)
  • [6] Palar P.S., Tsuchiya T., Parks G.T., A comparative study of local search within a surrogate-assisted multi-objective memetic algorithm framework for expensive problems, Applied Soft Computing Journal, 43, pp. 1-19, (2016)
  • [7] Rosales-Perez A., Coello C.A.C., Gonzalez J.A., Reyes-Garcia C.A., Escalante H.J., A hybrid surrogate-based approach for evolutionary multi-objective optimization, Proceedings of the 2013 IEEE Congress on Evolutionary Computation (CEC), pp. 2548-2555, (2013)
  • [8] Akhtar T., Shoemaker C.A., Multi objective optimization of computationally expensive multi-modal functions with RBF surrogates and multi-rule selection, Journal of Global Optimization, 64, 1, pp. 17-32, (2016)
  • [9] Montemayor-Garcia G., Toscano-Pulido G., A study of surrogate models for their use in multiobjective evolutionary algorithms, Proceedings of the 8th International Conference on Electrical Engineering Computing Science and Automatic Control (CCE), pp. 1-6, (2011)
  • [10] Guo G.Q., Li W., Yang B., Li W.B., Yin C., Predicting Pareto dominance in multi-objective optimization using pattern recognition, Proceedings of the 2nd International Conference on Intelligent System Design and Engineering Application (ISDEA), pp. 456-459, (2012)
12