Hybrid learning algorithm for Pi-sigma neural network and analysis of its convergence

doi:10.3778/j.issn.1002-8331.2008.35.017

Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (35): 56-58.DOI: 10.3778/j.issn.1002-8331.2008.35.017

• 理论研究 • Previous Articles Next Articles

Hybrid learning algorithm for Pi-sigma neural network and analysis of its convergence

NIE Yong,DENG Wei

College of Computer Science，Suzhou University of Science and Technology，Suzhou，Jiangsu 215006，China

Received:2007-12-25 Revised:2008-03-03 Online:2008-12-11 Published:2008-12-11
Contact: NIE Yong

Pi-sigma神经网络混合学习算法及收敛性分析

聂永,邓伟

苏州大学计算机科学与技术学院，江苏苏州 215006

通讯作者: 聂永

Abstract

Abstract: This paper uses a hybrid genetic algorithm to training Pi-sigma neural network and this algorithm is once applied to resolve a function optimizing problem.The hybrid genetic algorithm incorporates the stronger global search of genetic algorithm into the stronger local search of simplex method，and can search out the global optimum faster than genetic algorithm.The experiments show that the hybrid genetic algorithm can achieve better performance.At last，the hybrid genetic algorithm is proved converge to the global optimum with the probability of 1.

Key words: hybrid genetic algorithm, Pi-sigma neural network, algorithm convergence

摘要： 将一种解决函数优化问题的混合遗传算法用于Pi-sigma神经网络的训练。这种混合算法充分利用遗传算法算法的全局搜索能力，又利用了单纯型法的局部搜索能力，因此该混合遗传算法可以使Pi-sigma神经网络更快的收敛到全局最优解，而且收敛速度比遗传算法更快。实验证明了这种算法的优越性。最后还证明了该算法可以以概率1收敛到全局最优解。

关键词: 混合遗传算法, Pi-sigma神经网络, 算法收敛性

NIE Yong,DENG Wei. Hybrid learning algorithm for Pi-sigma neural network and analysis of its convergence[J]. Computer Engineering and Applications, 2008, 44(35): 56-58.

聂永,邓伟. Pi-sigma神经网络混合学习算法及收敛性分析[J]. 计算机工程与应用, 2008, 44(35): 56-58.

[1]	LIU Lanfen, YANG Xinfeng. Searching Algorithm for Efficient Paths Based on Hybrid Genetic Algorithm [J]. Computer Engineering and Applications, 2019, 55(11): 244-249.
[2]	WEI Jiangxia, CHEN Tian’en, ZHANG Chi. Improved hybrid genetic algorithm and its application research to optimize agricultural distribution [J]. Computer Engineering and Applications, 2016, 52(2): 234-238.
[3]	ZHANG Henan, ZHANG Shaowen. Improved hybrid genetic algorithm for university timetabling problem [J]. Computer Engineering and Applications, 2015, 51(5): 240-246.
[4]	YOU Wei, LEI Dingyou, ZHU Xiang. Biased random-key hybrid genetic algorithm for three-dimensional loading problem [J]. Computer Engineering and Applications, 2014, 50(22): 265-270.
[5]	FU Hao, GE Hongwei, SHAO Changlu, ZHU Liang. Models and hybrid genetic algorithm for scheduling shipbuilding inspectors [J]. Computer Engineering and Applications, 2014, 50(21): 238-242.
[6]	YU Xin, DENG Fei. Gradient algorithm with penalty for training recurrent pi-sigma neural network [J]. Computer Engineering and Applications, 2013, 49(4): 43-46.
[7]	LIU Bo1，2, LI Bo1, LI Yan1. Robust bilevel optimization model for emergency resource location and allocation under uncertain condition [J]. Computer Engineering and Applications, 2013, 49(16): 13-17.
[8]	LIU Fei，LV Shihui，ZHAO Zhonghua. Study of hybrid genetic algorithm based on multi-step reinforcement mutation operator [J]. Computer Engineering and Applications, 2011, 47(29): 46-48.
[9]	WANG Jianhua1，2，LI Nan2，GUO Hui2. Hybrid genetic algorithm for agile supply chain scheduling optimization [J]. Computer Engineering and Applications, 2011, 47(17): 224-228.
[10]	LI Nan1，2，GAO Pengdong1，LU Yongquan1，YU Wenhua1. Application research of parallel hybrid genetic algorithm in range images registration [J]. Computer Engineering and Applications, 2011, 47(12): 12-15.
[11]	LEI Yang，LI Shu-rong，ZHANG Qiang，ZHANG Xiao-dong. Novel hybrid GA based on bare bones PSO and its application [J]. Computer Engineering and Applications, 2010, 46(36): 7-10.
[12]	DU Zong-zong，LIU Guo-dong. Solution of TSP problem based on hybrid genetic simulated annealing algorithm [J]. Computer Engineering and Applications, 2010, 46(29): 40-42.
[13]	RONG Li-xia. Fuzzy simulation of vehicle routing problem under condition of fuzzy demand [J]. Computer Engineering and Applications, 2010, 46(18): 209-210.
[14]	LIU Feng¹，ZUO Xiao-jing²，ZHANG Kun¹. Novel genetic algorithm [J]. Computer Engineering and Applications, 2010, 46(12): 46-49.
[15]	YU Wu-yang. Hybrid genetic algorithm for transportation problem with multimodal transport [J]. Computer Engineering and Applications, 2009, 45(33): 10-12.

Hybrid learning algorithm for Pi-sigma neural network and analysis of its convergence

Pi-sigma神经网络混合学习算法及收敛性分析

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics