What is the role of the full connection layer?

Click on the above “ Xiaobai studies vision ”, Optional plus " Star standard " or “ Roof placement ”

Heavy dry goods , First time delivery

come from |  You know

author  |  Wei Xiushen

Address | https://www.zhihu.com/question/41037974/answer/150522307

edit |  Machine learning algorithm and Natural Language Processing official account

This article is only for academic sharing , If infringement , Please contact the background to delete

What is the use of the full connection layer ？ Let me talk about three points .

• Fully connected layer （fully connected layers,FC） In the whole convolution neural network “ classifier ” The role of . If we talk about convolution 、 The operation of pooling layer and activation function layer is to map the original data to the hidden layer feature space , The full connection layer plays the role of “ Distributed feature representation ” The role of mapping to the sample marker space . In practical use , The full connection layer can be realized by convolution operation ： If the front layer is fully connected, the fully connected layer can be transformed into convolution kernel 1x1 Convolution of ; The front layer is the full connection layer of the convolution layer, which can be transformed into convolution kernel hxw Global convolution of ,h and w They are the height and width of the convolution result of the previous layer （ notes 1）.

• At present, due to the parameter redundancy of the whole connection layer （ Only the full connection layer parameters can account for the whole network parameters 80% about ）, Recently, some network models with excellent performance, such as ResNet and GoogLeNet All of them are pooled by global average （global average pooling,GAP） replace FC To integrate the depth features learned , Finally still use softmax The equal loss function is used as the network objective function to guide the learning process . It's important to point out that , use GAP replace FC The network usually has better prediction performance . For specific cases, please refer to our ECCV 16（ video ） The practice of winning the champion in the image character analysis competition ：「 The way of champion 」Apparent Personality Analysis Competition experience sharing - Know about columns ,project：Deep Bimodal Regression for Apparent Personality Analysis

• stay FC At a time when people are less and less optimistic , Our recent research （In Defense of Fully Connected Layers in Visual Representation Transfer） Find out ,FC It can be used in the process of model representation capability migration “ A firewall ” The role of . In particular , Suppose that ImageNet The model obtained from the previous pre training is  , be ImageNet Can be regarded as the source domain （ In transfer learning source domain）. fine-tuning （fine tuning） It is the most commonly used transfer learning technology in the field of deep learning . For fine tuning , If the target domain （target domain） The image in is very different from the image in the source domain （ As compared ImageNet, The target domain image is not an object centered image , But the scenery , See the picture below ）, Not included FC The result of network fine-tuning is worse than that of including FC Network of . therefore FC Can be regarded as a model to represent the ability of “ A firewall ”, Especially when the source domain is quite different from the target domain ,FC Large models can be maintained capacity So as to ensure the transfer of model representation ability .（ Redundant parameters are not useless .）

notes 1: About convolution operation “ Realization ” Fully connected layer , It's necessary to be wordy .

With VGG-16 For example , Yes 224x224x3 The input of , The output of the final convolution layer is 7x7x512, If the back layer is a layer containing 4096 One neuron FC, Then the convolution kernel can be 7x7x512x4096 Global convolution to achieve this full connection operation , The convolution kernel parameters are as follows ：

“filter size = 7, padding = 0, stride = 1, D_in = 512, D_out = 4096”

After this convolution operation, the output is 1x1x4096.

If you need to stack another 2048 Of FC, Then the parameter can be set to “filter size = 1, padding = 0, stride = 1, D_in = 4096, D_out = 2048” Convolution layer operation .

Produced by Xiaobai team ： Zero basis proficient in semantic segmentation ↓↓↓

download 1：OpenCV-Contrib Chinese version of extension module

stay 「 Xiaobai studies vision 」 Official account back office reply ： Extension module Chinese course , You can download the first copy of the whole network OpenCV Extension module tutorial Chinese version , cover Expansion module installation 、SFM Algorithm 、 Stereo vision 、 Target tracking 、 Biological vision 、 Super resolution processing And more than 20 chapters .

download 2：Python Visual combat project 52 speak

stay 「 Xiaobai studies vision 」 Official account back office reply ：Python Visual combat project , You can download the Image segmentation 、 Mask detection 、 Lane line detection 、 Vehicle count 、 Add Eyeliner 、 License plate recognition 、 Character recognition 、 Emotional tests 、 Text content extraction 、 face recognition etc. 31 A visual combat project , Help fast school computer vision .

download 3：OpenCV Actual project 20 speak

stay 「 Xiaobai studies vision 」 Official account back office reply ：OpenCV Actual project 20 speak , You can download the 20 Based on OpenCV Realization 20 individual Actual project , Realization OpenCV Learn advanced .

Communication group

Welcome to join the official account reader group to communicate with your colleagues , There are SLAM、 3 d visual 、 sensor 、 Autopilot 、 Computational photography 、 testing 、 Division 、 distinguish 、 Medical imaging 、GAN、 Wechat groups such as algorithm competition （ It will be subdivided gradually in the future ）, Please scan the following micro signal clustering , remarks ：” nickname + School / company + Research direction “, for example ：” Zhang San  +  Shanghai Jiaotong University  +  Vision SLAM“. Please note... According to the format , Otherwise, it will not pass . After successful addition, they will be invited to relevant wechat groups according to the research direction . Do not Send ads within the group , Or you'll be invited out , Thanks for your understanding ~