Radial basis function

A radial basis function is a real-valued function that relies only on the distance from the origin, that is, φ (x) =φ (‖x‖), or the distance to any point C, where c is called the center point, which is φ (x,c) =φ (‖x-c‖). Any function that satisfies the φ (x) =φ (‖x‖) characteristic φ is called the radial basis function, the standard generally uses Euclidean distance (also called the European radial basis function), although other distance functions are also possible. In the neural network structure, it can be used as the main function of the full-join layer and the Relu layer.

Problems that require deep learning to solve problems have the following characteristics: There is a problem with the lack of depth. The human brain has a deep structure. The cognitive process is layered and gradually abstracted. Core IdeasEditThe core idea of deep learningThe learning structure as a network, the core idea of deep learning is as follows: ① unsupervised learning is used for each layer of network Pre-train;② each time with unsupervised learning only one layer, the training results as their first layer of input; ③ use a top-down monitoring algorithm to adjust all layer a). Autoencoder The simplest way is to use the characteristics of artificial neural networks, artificial neural Network (ANN) itself is a hierarchical structure of the system, if given a neural network, we assume that its output is the same as the input, and then train to adjust its parameters, to get the weights in each layer, naturally, We have several different representations of input I (each layer represents a representation), these representations are features, which can be found in the study, if the characteristics of the original features can be added to the automatic learning to improve the accuracy of the features, even in the classification of the best classification algorithm than the current effect is better! This method is called Autoencoder. Of course, we can continue to add some constraints to get the new deep learning method, such as: if Autoencoder based on the L1 regularity limit (L1 is mainly constrained in each layer of the nodes in most of the 0, only a few are not 0, This is the source of the sparse name, and we can get the sparse Autoencoder method. Because the input and output are the same, the output value of the middle-tier node represents the intermediate feature. This is the high-level abstraction of this input. Source: http://baike.baidu.com/link?url= Rh3o0p84n7tomoqli-hhvloxx3sqibr6pxg7sltnucqlwew733lols7ymx6vvvpvv3x1-0jbhlw2cxorsajvalnolhv0fppn-3lkqhg1xjdab5vhndg2qnvic Ykeuogc

Radial basis function