1. What are the features extracted by each layer of the convolutional neural network?
Convolutional neural network is a multi-layer neural network, each layer is composed of multiple two-dimensional planes, and each plane is composed of multiple independent neurons.
Figure: Demonstration of the concept of convolutional neural network: the input image is convolved with three trainable filters and offsets can be added. The filtering process is shown in Figure 1. After convolution, three feature maps are generated in the C1 layer, and then The four pixels in each group in the feature map are summed, weighted, and biased, and the feature maps of the three S2 layers are obtained through a Sigmoid function. These maps are further filtered to obtain the C3 layer. This hierarchy then produces S4 as S2 does. Finally, these pixel values are rasterized and concatenated into a vector input to a traditional neural network for output.
Generally, layer C is the feature extraction layer. The input of each neuron is connected to the local receptive field of the previous layer, and the local features are extracted. Once the local features are extracted, the positional relationship between it and other features is also Then it is determined; the S layer is the feature map layer, and each calculation layer of the network is composed of multiple feature maps, each feature map is a plane, and the weights of all neurons on the plane are equal. The feature map structure uses the sigmoid function with a small influence function kernel as the activation function of the convolutional network, so that the feature map has displacement invariance.
In addition, since neurons on a mapping surface share weights, the number of free parameters in the network is reduced, and the complexity of network parameter selection is reduced. Each feature extraction layer (C-layer) in the convolutional neural network is followed by a calculation layer (S-layer) for local averaging and secondary extraction. This unique two-time feature extraction structure makes the network It has a high distortion tolerance for input samples during recognition.
Google AI Writing Project: Little Fat Cat
2. About the feature extraction of one-dimensional signal by convolutional neural network
Hello, the feature extraction of the signal is actually a filtering operation from a mathematical point of view, and it is actually realized by convolution. Reference: a study note for a technical nerd . The following is a matlab implementation:
function r= my_conv(a, b)
m=length(a);
n=length(b);
r=zeros(1, m+n-1);
for k = 1:m
c = a(k)*b;
d = r(1, k:k+n-1);
d = d+c;
r(1, k:k+n-1) = d;
end
3. The neural network extracts the probability distribution characteristics of the image
The neural network extracts the probability distribution features of the image:
Since neurons on a mapping surface share weights, the number of free parameters in the network is reduced, and the complexity of network parameter selection is reduced. Each feature extraction layer (C-layer) in the convolutional neural network is followed by a calculation layer (S-layer) for local averaging and secondary extraction. This unique two-time feature extraction structure makes the network It has a high distortion tolerance for input samples during recognition.
Neural Network Features:
For example, when realizing image recognition, we only need to input many different image templates and corresponding recognition results into the artificial neural network, and the network will gradually learn to recognize similar images through self-learning function. The self-learning function is particularly important for forecasting. It is expected that artificial neural network computers in the future will provide human beings with economic forecasts, market forecasts, and benefit forecasts, and their application prospects are very promising.