In a world of increasingly accessible data, unsupervised learning algorithms are becoming more and more efficient and profitable. Companies that understand this will soon have a competitive advantage over those who are slow to jump on the artificial intelligence bandwagon. As a result, developers with Machine Learning and Deep Learning skills are increasingly in demand and have gold on their hands.

In this course, we will see how to take advantage of a raw dataset, without any labels. In particular, we will focus exclusively on Autoencoders and Variational Autoencoders and see how they can be trained in an unsupervised way, making them particularly attractive in the era of Big Data.

This course, taught using the Python programming language, requires basic programming skills. If you don't have the required foundation, I recommend that you brush up on your skills by taking a crash course in programming. Also, it is best to have basic knowledge of optimization (we will use gradient optimization) and machine learning.

Concepts covered:

• Autoencoders and their implementation in Python

• Variational Autoencoders and their implementations in Python

• Unsupervised Learning

• Generative models

• PyTorch through practice

• The implementation of a scientific ML paper (Auto-Encoding Variational Bayes)