Generative and Discriminative modelling

[bkhanal-11]

What is the difference between generative and discriminative modelling?

[bkhanal-11]

Generative modelling:

Generative modelling is a statistical modeling technique that aims to model the entire probability distribution of the data. In other words, it tries to model the underlying patterns and relationships that generate the observed data. Given a set of inputs, a generative model generates new examples that are similar to the training data. Generative models are trained to learn the joint probability distribution $P(X,Y)$
of the input $X$ and the output $Y$

Some examples of generative models include Gaussian mixture models, Hidden Markov Models, and Variational Autoencoders.

Discriminative modelling

Discriminative modelling, on the other hand, focuses on modeling the conditional probability distribution $P(Y|X)$
, which is the probability of the output $Y$ given the input $X$. Discriminative models try to learn the decision boundary that separates different classes in the data. Discriminative models are trained to learn the mapping function from input $X$ to output $Y$

Some examples of discriminative models include Logistic Regression, Support Vector Machines, and Neural Networks.

[KhanalBijay]

I found this helpful too.