Read the blog link: https://blog.csdn.net/cherry_yu08/article/details/80846146
1. Understanding of the feeling field: https://blog.csdn.net/u010725283/article/details/78593410
https://blog.csdn.net/qq_36653505/article/details/83473943
(In layman's terms, each feature (each pixel) of the final output of the image is affected by which part of the original image.)
2. A blog addressing gradient disappearance and gradient explosion (but did not mention the intermediate supervision used in this article): https://blog.csdn.net/Sakura55/article/details/84571422
3. The improvement of the VGG network structure model compared to the Alexnet network model: https://blog.csdn.net/dcrmg/article/details/79254654
(Replace a large convolution kernel with a stack of multiple small convolution kernels)
Why use multiple small convolution kernels to replace the big one: https://blog.csdn.net/ytusdc/article/details/85265057