Neural Networks
Neural Networks are the power behind personalized recommendations. CNN or Convolutional Neural Network is a class of artificial neural networks. When we think of neural networks, we think of matrix multiplication, but CNN uses a mathematical technique called convolution.
In neural networks, in addition to data analysis algorithms, different types of algorithms such as First In First Out (FIFO), Strict Priority (SP), Weighted Fair Queuing (WFQ) and Deficit Resolution (DRR) are used.
In computer language, a neuron can be defined as a function that receives the outputs of other “neurons” and transmits a number between 0 and 1. Thus, artificial neurons are connected to each other to form an artificial neural network. Each neural network has an input and output consisting of a large number of nodes. The neural network receives information through the input layer, while the output layer represents the final results. There are several stages (columns of neurons) between the input and output. These are called hidden layers and their main function is to transform information into smaller pieces that are easier to process. And because there can be many hidden layers, the algorithm is called deep learning.
CNN
A Convolutional Neural Network (CNN) represents a type of artificial neural network; it is also one of the best known deep neural networks. This class of deep neural networks is often used for image analysis, recognition and processing of pixel data. CNNs use deep learning algorithms to complete descriptive and generative tasks. They also occasionally apply image and video recognition, recommender systems and natural language processing (NLP).
GNN
GNNs are Graph Neural Networks, which represent a class of deep learning algorithms. As the name suggests, they are often applied in graph analysis (in Computer Science, the term “graph” is understood as a data structure consisting of vertices (often called nodes) and edges, or sets of vertices and edges).
The typical task of GNNs is node classification. A typical node has two properties: its intrinsic property and its ground truth label (which helps to check whether the machine learning results are in line with the real world). In a partially labeled graph, GNNs use the labeled nodes, analyze them and then predict the likely label in the unlabeled ones. In general, Graph Neural Networks are used for the following:
– Node Classification
– Graph Classification
– Graph Visualization
– Connection Forecast
– Graph Clustering
GAN
Generative Adversarial Networks (GANs) represent a class of machine learning methods that use two neural networks and “favor one over the other” to improve the accuracy of their predictions. This AI technique has given computers the human trait of creativity.
