竞争合作行为下的深度演化算法

doi:10.3778/j.issn.1673-9418.1905088

摘要/Abstract

摘要：

将深度与演化算法结合，提出一种深度演化算法，即群竞争合作优化（GCCO）算法。首先引入生物群模型来模拟群体搜索猎物的自然现象，算法通过多步迭代可简单实现优化问题求解。在生物群模型中，跟随者采用变步长的区域复制方式，平衡了收敛速度与优化精度，随机者采用基于特征变换的随机游走模式，避免陷入局部最优。其次引入竞争模型和合作模型增加算法复杂性，通过群体间的竞争和信息共享，提高算法的搜索性能。算法的数学模型是从群论、动力学以及帝国竞争理论推导出来的，在理论上也分析验证了算法的收敛性。最后在十个优化基准函数上与其他三种优化算法对比测试算法的性能。在解决上海市设立燃气站点提高到场及时率的实际问题中，GCCO算法也取得了比其他算法更好的效果。

关键词: 深度演化, 特征变换, 竞争模型, 合作模型

Abstract:

Combining depth with evolutionary algorithms, a deep evolutionary algorithm, the group competition cooperation optimization (GCCO) algorithm, is proposed. Firstly, the bio-group model is introduced to simulate the natural phenomenon that groups search prey. The algorithm can easily solve the optimization problem through multi-step iteration. In bio-group model, the follower adopts a variable step size region replication method to balance the convergence speed and optimization precision. The wanderer adopts random walk mode based on feature transfor-mation to avoid local optimization. Secondly, the introduction of competition model and cooperation model increases the complexity of the algorithm, and improves the search performance of the algorithm through competition and information sharing among groups. In addition, the mathematical model of the algorithm is derived from group theory, dynamics and imperial competition theory. The convergence of the algorithm is also verified theoretically. Finally, the performance of the proposed algorithm is tested through comparing it with the other three optimization algorithms on ten optimization benchmark functions. At the same time, GCCO achieves better results than other algorithms in setting up gas stations in Shanghai to improve the on-time rate.

Key words: deep evolution, feature transformation, competition model, cooperation model

陈海娟，冯翔，虞慧群. 竞争合作行为下的深度演化算法[J]. 计算机科学与探索, 2020, 14(7): 1114-1125.

CHEN Haijuan, FENG Xiang, YU Huiqun. Deep Evolution Algorithm Under Competitive and Cooperative Behavior[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(7): 1114-1125.

参考文献

[1] Zhou Z H, Feng J. Deep forest: towards an alternative to deep neural networks[C]//Proceedings of the 26th Interna-tional Joint Conference on Artificial Intelligence, Melbourne, Aug 19-25, 2017: 3553-3559.
[2] Lecun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(7553): 436.
[3] Schmidhuber J. Deep learning in neural networks: an over-view[J]. Neural Networks, 2015, 61: 85-117.
[4] Simon D. Biogeography-based optimization[J]. IEEE Transac-tions on Evolutionary Computation, 2008, 12(6): 702-713.
[5] Cuevas E, Cienfuegos M, Zaldívar D, et al. A swarm opti-mization algorithm inspired in the behavior of the social-spider[J]. Expert Systems with Applications, 2013, 40(16): 6374-6384.
[6] Alizadeh E, Meskin N, Khorasani K. A dendritic cell immune system inspired scheme for sensor fault detection and isola-tion of wind turbines[J]. IEEE Transactions on Industrial In-formatics, 2018, 14(2): 545-555.
[7] Feng X, Wang Y B, Yu H Q, et al. A novel intelligence algo-rithm based on the social group optimization behaviors[J]. IEEE Transactions on Systems, Man, and Cybernetics: Sys-tems, 2018, 48(1): 65-76.
[8] Eckardt W. Cooperation and competition in two forest mon-keys[J]. Behavioral Ecology, 2017, 15(3): 400-411.
[9] Fehr E, Schmidt K M. A theory of fairness, competition and cooperation[J]. The Quarterly Journal of Economics, 1999, 114(3): 817-868.
[10] Atashpaz-Gargari E, Lucas C. Imperialist competitive algo-rithm: an algorithm for optimization inspired by imperialistic competition[C]//Proceedings of the 2007 IEEE Congress on Evolutionary Computation, Singapore, Sep 25-28, 2007. Pis-cataway: IEEE, 2007: 4661-466.
[11] West S A, Griffin A S, Gardner A. Evolutionary explanations for cooperation[J]. Current Biology, 2007, 17(16): R661-R672.
[12] Heg D, Van Treuren R. Female-female cooperation in polyg-ynous oystercatchers[J]. Nature, 2017, 391: 687-691.
[13] Shi B B, Chen Y N, Zhang P, et al. Nonlinear feature trans-formation and deep fusion for Alzheimer??s disease staging analysis[J]. Pattern Recognition, 2017, 63: 487-498.
[14] Desai A, Lee D J. Visual odometry drift reduction using SYBA descriptor and feature transformation[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(7): 1839-1851.
[15] Zhu X H, Ni Z W, Cheng M Y. Self-adaptive improved artificial fish swarm algorithm with changing step[J]. Com-puter Science, 2015, 42(2): 210-216. 朱旭辉, 倪志伟, 程美英. 变步长自适应的改进人工鱼群算法[J]. 计算机科学, 2015, 42(2): 210-216.
[16] Wang L, Zhong X, Liu M. A novel group search optimizer for multi-objective optimization[J]. Expert Systems with App-lications, 2012, 39(3): 2939-2946.
[17] Li Y Z, Zheng X W, Xiao X C. A study on cooperative multi-objective group search optimizer[C]//Proceedings of the 27th Chinese Control and Decision Conference. Piscataway: IEEE, 2015: 3776-3781.
[18] He S, Wu Q H, Saunders J R. Group search optimizer: an optimization algorithm inspired by animal searching beha-vior[J]. IEEE Transactions on Evolutionary Computation, 2009, 13(5): 973-990.
[19] Peng H, Chen H Y. A new variable step size LMS algorithm and its analysis[J]. Journal of Zhejiang University of Tech-nology, 2018, 46(1): 45-50. 彭宏, 陈泓宇. 一种新的变步长LMS算法及分析[J]. 浙江工业大学学报, 2018, 46(1): 45-50.
[20] Derrac J, García S, Molina D, et al. A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms[J]. Swarm & Evolutionary Computation, 2011, 1(1): 3-18.
[21] García S, Fernández A, Luengo J, et al. A study of statistical techniques and performance measures for genetics-based mac-hine learning: accuracy and interpretability[J]. Soft Computing, 2009, 13(10): 959-977.