It also solves the problem how to decide the hiddennode Numbers for the BP neural networks.
同时，解决了BP神经网络难以确定隐结点数的问题。
2
Furthermore, an annealing robust learning algorithm is presented to adjust these hiddennode parameters as well as the weights of the SVR-NN.
此外，退火的鲁棒学习算法，提出这些隐藏的节点参数以及SVR - NN的权重的调整。
3
where d(n) for a hiddennode n, turns on (1) how much n influences any given output node; and (2) how much that output node itself influences the overall error of the net.
其中 d(n) 是隐藏节点 n 的函数，让我们来看（1）n 对任何给出的输出节点有多大影响；（2）输出节点本身对网络整体的误差有多少影响。