The generative model must understand the distribution from which the data is obtained and must then use this understanding to perform the classification task.
Generative models can create data similar to the training data they received because they learned from the distribution from which the data is provided.
Using this information, it can compare attributes to classify an image in the same way that the discriminating algorithm can classify that image.
This can generate a new image that looks like one of the class images provided to it for training.
Training a generative model starts with collecting a huge amount of data in a given domain and then training a model to generate similar data.
Welcome, my name is Eric Soupet and I am the administrator of the site elodees.com. elodees.com is a state of the art of Artificial Intelligence and aims to be collaborative, you can now offer content such as articles, events, tutorials, ... so don't hesitate !
