Baidu Encyclopedia version

Self-organizing neural network SOM is an important type of neural network based on unsupervised learning methods. The theory of self-organizing mapping networks was first proposed by X. XhenX, Helsinki University of Technology, Finland. Since then, along with the rapid development of neural networks in the 1981 century 20 era, the self-organizing mapping theory and its applications have also made great progress.

see details

 

Wikipedia version

Self-Organizing Map (SOM) or Self-Organizing Feature Mapping (SOFM) is a type of artificial neural network (ANN) that uses trained unsupervised learning to produce low-dimensional (usually two-dimensional), discrete representation training samples. The input space, called the map, is a way to reduce the number of dimensions. Self-organizing maps are different from other artificial neural networks in that they apply competing learning rather than error correcting learning (for example, backpropagation with gradient descent), in the sense that they use neighborhood functions to preserve the topology of the input space. Attributes.

这使得SOM 通过创建高维数据的低维视图(类似于多维缩放)对可视化非常有用。芬兰教授Teuvo Kohonen在20世纪80年代引入的人工神经网络有时被称为Kohonen地图或网络。Kohonen网是一种计算上方便的抽象,建立在20世纪70年代神经系统的生物模型上和形态发生模型可追溯到20世纪50年代的阿兰图灵。

see details