Restricted Boltzmann Machines - A Friendly Introduction

Dive into a friendly introduction to Restricted Boltzmann Machines (RBM) and their training process through this 37-minute video. Explore the concept using a real-life example involving people and pets, starting with an intriguing mystery. Learn about scoring systems, probability calculations, and the training process, including contrastive divergence and Gibbs sampling. Understand how to update weights, tackle sampling problems, and perform independent sampling. Discover techniques for picking random samples with conditions and completely random samples. Conclude with a comprehensive summary that ties all the concepts together, providing a solid foundation for understanding RBMs and their applications in machine learning.

Introduction:
Mystery:
Scores:
Probabilities:
Training
Contrastive Divergence:
Small Problem:
Gibbs Sampling:
Updating Weights:
Sampling Problems:
Independent Sampling:
Picking Random Samples with Conditions:
Picking Completely Random Samples:
Summary:
Conclusion: