Chapter 14 (focus on Sections 14.1–2, 14.4–5)

1. Describe the structure and semantics of Bayesian networks.
2. Describe the inference by enumeration algorithm and compare and contrast it with the use of the full joint distribution.
3. Compare and contrast exact inference in Bayesian networks with approximate inference using Rejection Sampling, Likelihood Weighting and Gibbs Sampling.

For a detailed introduction to Bayes Networks, see the Udacity AI materials (Lesson 4, “Probability in AI”, S. Thrun, link on the policies page, Episodes 31–34, “General Bayes Net”).

Programming Tools

1. Scikit-learn — Note the range of pre-built algorithms provided by SciKit-Learn.

   • An introduction to machine learning with scikit-learn: Read through this tutorial, running the code that creates and trains a Support-Vector Classification (SVC) model on the well-known Iris dataset.

   In this course, we will generally focus on deep learning models, but it’s important to realize the breadth of learning algorithms provided by SciKit-Learn.