Erb R J
Clinical Research Foundation-America, Lenexa, Kansas 66219.
Pharm Res. 1993 Feb;10(2):165-70. doi: 10.1023/a:1018966222807.
Neurocomputing is computer modeling based, in part, upon simulation of the structure and function of the brain. Neural networks excel in pattern recognition, that is, the ability to recognize a set of previously learned data. Although their use is rapidly growing in engineering, they are new to the pharmaceutical community. This article introduces neurocomputing using the backpropagation network (BPN).
神经计算部分基于对大脑结构和功能的模拟进行计算机建模。神经网络在模式识别方面表现出色,即识别一组先前学习的数据的能力。尽管它们在工程领域的应用正在迅速增长,但在制药领域却是新生事物。本文介绍了使用反向传播网络(BPN)的神经计算。