What is graph neural network? Figure what is the use of neural networks?

Deep learning is the core technology in the field of artificial intelligence in recent years , With its further development , The rise and landing application of artificial intelligence today . When it comes to deep learning , We have to mention graph neural network (GNN), After all, graph neural network is the foundation of deep learning technology . Next, I will take you three minutes to interpret the graph neural network , The main content includes the definition of graph neural network 、 Rise and use .

What is a graph neural network ?

1、 Figure definition of neural network

In recent years , Figure the rise and application of neural network has successfully promoted the research of artificial intelligence in pattern recognition and data mining .

GNN Full name graph neural network , there G Yes yes (Graph) It means ,GNN Why is it important , Because the graph is very important . Graph is a very important data structure in computer science , Computer science has a compulsory basic course called “ Discrete Mathematics ”, It sounds like a branch of mathematics , But after all “ Discrete Mathematics ” Where is the boundary of , There is no unified conclusion yet . But there is a point of knowledge , All versions of 《 Discrete Mathematics 》 No textbook will be missed , That's it “ graph theory ”, Discuss a kind of call “ chart ” Data structure of . and GNN Inside “ chart ”, It refers to graph theory “ chart ”.

So what exactly is “ chart ”? Just two things , The vertices (Vertex) He Bian (Edge). The so-called vertex , The nodes in the network topology , For example, in the network topology PC machine 、 Servers and routers, etc , And the so-called edge , Is the line connecting these network nodes . So graph is widely used , Network topology is a very typical graph structure .

2、 Figure the rise of neural networks

Figure the emergence of neural network is essentially the rise of a new technology , So why launch this new technology ? Launch a new technology , The subtext is that the original technology has shortcomings , Let's take a look at CNN and RNN The inadequacies of being . To put it bluntly, it is the data structure , Models feed data , We all know that . But the existing deep learning model , Whether it's CNN, still RNN, Or something else , Both have a requirement for the data structure of the data , Must all be Euclidean structures . What looks square is Euclidean structure , Military training parade square , The horizontal and vertical directions are one person next to another , This is the typical Euclidean structure . Graph is a non Euclidean structure , So there is no way to use the traditional depth model . therefore , Researchers have developed graphical neural networks .

3、 Figure application of neural network

In recent years , Deep learning brings face recognition 、 The successful application of voice assistant and machinetranslation . The three scenarios represent three types of data ： Images 、 Voice and text . The key to the breakthrough of deep learning in these three scenarios is the end-to-end learning mechanism behind it . in addition , The industry believes that large-scale graph neural network is a powerful reasoning method for cognitive intelligent computing . Graph neural network extends deep neural network from processing traditional unstructured data to higher-level structured data . More Than This , Graph also has strong semantic visualization ability , This advantage is shared by all GNN Shared by the model . For example, in the scenario of abnormal transaction account identification ,GNN After judging an account as an abnormal account , The partial subgraph of the account can be visualized .

Share some of my artificial intelligence learning materials for free , Including some AI Common framework actual combat video 、 Image recognition 、OpenCV、NLQ、 machine learning 、pytorch、 Computer vision 、 Videos such as deep learning and neural network 、 Courseware source code 、 Famous essence resources at home and abroad 、AI Hot papers 、 Industry reports, etc .

For better systematic learning AI, I recommend that you collect one .

Here are some screenshots , Free download method is attached at the end of the article .

One 、 Artificial intelligence courses and projects

Two 、 Famous essence resources at home and abroad

3、 ... and 、 Collection of papers on artificial intelligence

Four 、 AI Industry Report

Learn Artificial Intelligence well , Read more , Do more , practice , If you want to improve your level , We must learn to settle down and learn slowly and systematically , Only in the end can we gain something .

Click on the business card below , Scan the code and download the information for free .