Part 5 - Boltzmann Machines
Welcome to Part 5 - Boltzmann Machines
- In this part you will learn:
- The Intuition of Boltzmann Machines
- How to build a Boltzmann Machine
- Boltzmann Machines can be seen from two different points of view:
- An Energy-Based Model
- A Probabilistic Graphical Model
In the Intuition Lectures Kirill will focus more on the Energy-Based Model point of view, and then for the Practical Lectures we will focus more on the Probabilistic Graphical Model point of view.
- In these last two parts (Part 5 - Boltzmann Machines and Part 6 - AutoEncoders) of this course, we will create two types of Recommender Systems:
- One that predicts binary ratings “Like” or “Not Like”. We will build it in this Part 5 with a Boltzmann Machine.
- Another one that predicts ratings from 1 to 5. We will build it in Part 6 with an AutoEncoder.
- We will implement these two Deep Learning models with PyTorch, a highly advanced Deep Learning platform more powerful than Keras. Every single line of code will be explained in details but I would recommend to have a first look at the PyTorch documentation to start getting familiar with PyTorch:
Enjoy Deep Learning!
Plan of Attack
- What we will learn in this section:
- The Boltzmann Machine
- Energy-Based Models (EBM)
- Restricted Boltzmann Machine (RBM)
- Contrastive Divergence (CD)
- Deep Belief Networks (DBN)
Read More about Boltzmann Machines»
Download Resources