The Most Common Form of Image Identification Training

The most common form of training, well, in this setup is so-called supervised training where you say, here's a picture, it's a cat. You do that in typical image identification training in the early days of this, wouldn't use like 25 million images of a total of maybe 5,000 kinds of objects. So what you want to do is kind of bash the neural net. You want to run that enough times with enough different pieces of text and sort ofbash the neural net hard enough that it will eventually learn correctly to predict. Now the whole thing that's exciting with something like chat GPT or so-called generative AI is that it is often being given inputs