Machine Learning:Neural Network
1,Wikipedia上对Neural Network的定义:
In machine learning,artificial neural networks (ANNs) are a family of statistical learning algorithms inspired by biological neural networks (thecentral nervous systems of animals, in particular thebrain) and are used to estimate or approximate functions that can depend on a large number of input sand are generally unknown. Artificial neural networks are generally presented as systems of interconnected "neurons" which can compute values from inputs, and are capable ofmachine learning as well aspattern recognition thanks to their adaptive nature.
2,为什么要引入Neural Network?
我们前面学习了回归和分类模型,然而它们的实际应用受制于数据的规模(也就是维度灾难(curse of dimensionality))。Neural Network用于处理大量输入特征是非常有优势的,例如抽取计算机视觉中图片的像素作为输入特征,那么这将得到数量巨大的输入特征集;如果仍然回归和分类模型,学习参数所需要的时间将会是不能承受的;
二:Neural Network—Representation
1,Neural Network模型
在神经网络中,我们把第一层称为input layer,最后一层称为output layer,中间若干层都称为hidden layer。
下面我们来看一个非常简单的神经网络: