Generative models are a class of Deep Learning models that learn the underlying distribution of training data. There are diverse generative models depending on how the model distribution is represented (explicitly or implicitly through sampling) and how the learning objective is formulated. In this talk, we introduce the concept of generative models and their several examples, such as Variational Autoencoder (VAE), Generative Adversarial Network (GAN), and Optimal Transport Map. If time permits, we would also introduce the Energy-based Models as an example of explicit generative model.