Page 327 - Ai_417_V3.0_C9_Flipbook
P. 327
Input Image
Label: Dogs
Label: Cats
Output
Cats = 2 Dog = 1
In supervised learning, discriminative modelling contrasts with generative modelling, where the goal is to model
the joint probability distribution of both the input features and the output labels. Generative models can be
used to generate new data points that resemble the training data, whereas discriminative models are primarily
focused on classification or regression tasks.
Let us consider an example.
A father has two kids, Kid A and Kid B. Kid A has a special learning ability where he can learn everything in depth.
Kid B has a special learning ability where he can only learn the differences between what he saw.
One fine day, the father shows his kids (Kid A and Kid B) two kinds of animals, a dog and a cat. After a few days,
the father showed them an animal and asked both of them “Is this animal a dog or a cat?”
Kid A drew the image of a dog and a cat on a piece of paper based on what he saw earlier. He compared both the
images with the animal standing before him and answered based on the closest match of the image and animal,
he answered: “The animal is a dog.” Kid B knows only the differences, based on different properties learned, he
answered: “The animal is a dog.”
Here, we can see both of them are finding the kind of animal, but the way of learning and the way of finding
the answer is entirely different. In machine learning, we generally call Kid A as a generative model & Kid B as a
discriminative model.
Introduction to Generative AI 325

