The automatic model selection and variable kernel width for RBF neural networks
The automatic model selection and variable kernel width for RBF neural networks
复制标题
DOI：
10.1016/j.neucom.2011.07.011
复制
发表时间：
2011-10
期刊：
影响因子：
6
通讯作者：
Peng Zhou;Dehua Li;Hong Wu;Feng Cheng
中科院分区：
文献类型：
--
作者：
Peng Zhou;Dehua Li;Hong Wu;Feng Cheng
正交最小二乘（OLS）算法已广泛应用于RBF网络的基选择，但由于必须手动指定容差ρ，因此无法自动执行模型选择。这会引入噪声，并且在参数复杂的实时系统中很难实现。因此，提出了一种检测其基函数的最佳数量的通用标准。本文不仅将用于适应度计算的贝叶斯信息准则（BIC）方法纳入OLS算法的基函数选择过程中以分配合适的数量，而且还提出了一种新的方法来优化高斯函数的宽度，以提高泛化性能。将增强算法应用于已知和未知噪声非线性动态系统的径向基函数神经网络（RBFNN），并与标准OLS进行性能比较；实验结果表明，BIC对于适应度计算的功效以及正确选择基函数宽度的重要性都是显着的。
The Orthogonal Least Squares (OLS) algorithm has been extensively used in basis selection for RBF networks, but it is unable to perform model selection automatically because the tolerance ρ must be specified manually. This introduces noise and it is difficult to implement in the parametric complexity of real-time system. Therefore, a generic criterion that detects the optimum number of its basis functions is proposed. In this paper, not only the Bayesian Information Criterion (BIC) method, used for fitness calculation, is incorporated into the basis function selection process of the OLS algorithm for assigning its appropriate number, but also a new method is developed to optimize the widths of the Gaussian functions in order to improve the generalization performance. The augmented algorithm is employed to the Radial Basis Function Neural Networks (RBFNN) for known and unknown noise nonlinear dynamic systems and its performance is compared with the standard OLS; experimental results show that both the efficacy of BIC for fitness calculation and the importance of proper choice of basis function widths are significant.