针对BP神经网络易陷入局部极小点、泛化能力低的缺陷,提出了BP神经网络的IAP-SOBPNN(Particle Swarm Optimization with Immunity Algorithm Back Propagation Neural Network)组合训练算法,即免疫算法、粒子群算法和BP算法的组合.将此...针对BP神经网络易陷入局部极小点、泛化能力低的缺陷,提出了BP神经网络的IAP-SOBPNN(Particle Swarm Optimization with Immunity Algorithm Back Propagation Neural Network)组合训练算法,即免疫算法、粒子群算法和BP算法的组合.将此组合训练算法应用到非线性函数逼近和具有复杂非线性动力学特征的股价预测中,仿真实验表明,该算法避免了网络陷入局部极小点,提高了网络的泛化能力,同时为BP网络参数的确定提供了一条崭新的思路.展开更多
Designing a fuzzy inference system(FIS)from data can be divided into two main phases:structure identification and parameter optimization.First,starting from a simple initial topology,the membership functions and syste...Designing a fuzzy inference system(FIS)from data can be divided into two main phases:structure identification and parameter optimization.First,starting from a simple initial topology,the membership functions and system rules are defined as specific structures.Second,to speed up the convergence of the learning algorithm and lighten the oscillation,an improved descent method for FIS generation is developed.Furthermore, the convergence and the oscillation of the algorithm are system- atically analyzed.Third,using the information obtained from the previous phase,it can be decided in which region of the in- put space the density of fuzzy rules should be enhanced and for which variable the number of fuzzy sets that used to partition the domain must be increased.Consequently,this produces a new and more appropriate structure.Finally,the proposed method is applied to the problem of nonlinear function approximation.展开更多
TP381 2001053318混合神经网络及其在非线性系统控制中的应用=Mixed neural network and itS application to the control of nonlinear sysetms[刊,中]/安凯(中科院光电技术研究所.四川,成都(610209))//光电工程.—2000,27(5).—1-4针...TP381 2001053318混合神经网络及其在非线性系统控制中的应用=Mixed neural network and itS application to the control of nonlinear sysetms[刊,中]/安凯(中科院光电技术研究所.四川,成都(610209))//光电工程.—2000,27(5).—1-4针对一类非线性动态系统模型的特点。展开更多
文摘针对BP神经网络易陷入局部极小点、泛化能力低的缺陷,提出了BP神经网络的IAP-SOBPNN(Particle Swarm Optimization with Immunity Algorithm Back Propagation Neural Network)组合训练算法,即免疫算法、粒子群算法和BP算法的组合.将此组合训练算法应用到非线性函数逼近和具有复杂非线性动力学特征的股价预测中,仿真实验表明,该算法避免了网络陷入局部极小点,提高了网络的泛化能力,同时为BP网络参数的确定提供了一条崭新的思路.
基金Supported by National Basic Research Program of China(973 Program)(2007CB714006)
文摘Designing a fuzzy inference system(FIS)from data can be divided into two main phases:structure identification and parameter optimization.First,starting from a simple initial topology,the membership functions and system rules are defined as specific structures.Second,to speed up the convergence of the learning algorithm and lighten the oscillation,an improved descent method for FIS generation is developed.Furthermore, the convergence and the oscillation of the algorithm are system- atically analyzed.Third,using the information obtained from the previous phase,it can be decided in which region of the in- put space the density of fuzzy rules should be enhanced and for which variable the number of fuzzy sets that used to partition the domain must be increased.Consequently,this produces a new and more appropriate structure.Finally,the proposed method is applied to the problem of nonlinear function approximation.
文摘TP381 2001053318混合神经网络及其在非线性系统控制中的应用=Mixed neural network and itS application to the control of nonlinear sysetms[刊,中]/安凯(中科院光电技术研究所.四川,成都(610209))//光电工程.—2000,27(5).—1-4针对一类非线性动态系统模型的特点。
