Course: MIT 6.S191: Introduction to Deep Learning

Lecture video: https://www.youtube.com/watch?v=QcLlc9lj2hk&list=PLtBw6njQRU-rwp5__7C0oIVt26ZgjG9NI&index=4&ab_channel=AlexanderAmini

Lecturers: Alexander Amini and Ava Soleimany

Introduction to Generative Modeling

This lecture covers techniques and methods that looks at data and generates brand new data instances based on learned patterns from a model.

You may not be able to tell right away, but the following faces have been generated by a generative model train on large dataset with faces.

In the past lecture, we learned about supervised learning mostly, which covers models that learn a function that maps data to labels.

In contrast, unsupervised learning concerns with training models on data without labels and learning the underlying structure that defines the distribution of the data. In traditional machine learning, you may have seen unsupervised learning such as clustering or PCA.