论文部分内容阅读
采用随机梯度上升非监督学习规则的单层线性前馈神经网络可用于测井数据的主成分分析。本文阐明了这种方法的基本原理,并将由这种方法得到的结果与传统的统计学方法—K-L变换得到的结果做了比较。
A single linear feedforward neural network using unsupervised learning rules with stochastic gradient ascent can be used for principal component analysis of log data. This paper illustrates the basic principle of this method and compares the results obtained by this method with the results obtained by the traditional statistical method, the K-L transformation.