I'm super excited about this course! It appeared in my feed for 'machine learning'. This is a course from Ecole Polytechnique Federale de Lausanne and it is about to start in a couple of days.
It's a deep learning course taught by Francois Fleuret. The first lecture is on February 21, while the last one, Lecture 14 is on May 30, 2018. I need to mention that the lectures, materials and resources are free and available at the course page.
Students in this class are given a virtual machine to download to avoid the inconveniences of installing all the libraries and software required for work during the class. So, a virtual machine is basically a system that simulates an entire computer, in this case an operating system with all the libraries and resources preinstalled. As of such, everyone would be on the same page.
I personally use virtual machines not only for machine learning/deep learning, but also for training in offensive security and penetration testing - for this I use vulnerable machines.
Anyway, this course starts with some history and math concepts that are required in deep learning. Then it goes through:
- ML fundamentals (bias-variance trade-off, regression, PCA, K-means)
- Multi-Layer Perceptrons, gradient descent and backpropagation
- Convolutional Neural Nets
- Computer vision for image classification, object detection and semantic segmentation
- Autoencoders
- Generative Adversarial Nets
- Recurrent Nets
- And more.
Looking over a few of the slides, I gotta tell you they are pretty dense. Some decks are 200+ slides long, which means that concepts are being presented and explained in sufficient detail.
Each lecture has some practical exercises (most programming) and some of them have a mini-project. I suspect they might record the lectures for later view. If that happens, it would be awesome!
Now, this course is not for non-experts or the lay non-technical public. It is geared more towards those with good knowledge in the field. So, the prerequisites are: linear algebra, differential calculus, probabilities and statistics, programming, and more.
If you have this type of knowledge, you can jump right in and go through the first lecture as it's already available at the course page. I'd love to hear some preliminary feedback if you do that. I'd probably work my way through it starting the following days:
Deep Learning Course from Polytechnique Lausanne - [EE-559 2018]
To stay in touch with me, follow 
Cristi Vlad Self-Experimenter and Author