The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform. … Its province is to assist us in making available what we’re already acquainted with. —A. Lovelace, Sketch of the Analytical Engine, 1853

  1. Chapter 18 (focus on Sections 18.1–3)

    1. Compare and contrast the three main types of machine learning.
    2. Describe:
      1. The nature of Decision Trees
      2. The information theoretical concepts of: Entropy and Information Gain
      3. Overfitting

  2. Guides for the Google’s Machine Learning Crash Course require that you to work through the specified units, watching the videos, reviewing the textual materials (mostly reiterations of the videos), doing the “Check Your Understanding” and “Playground” exercises, all with a focus on the specific concepts listed in the guide. Note that we’ll save Google’s Problem Framing “Try it Yourself” exercises and Crash Course “Programming” exercises for the labs.

    1. Introduction to Machine Learning
      1. What are Norvig’s reasons for studying machine learning?
    2. Framing
      1. Know the key “ML Terminology”:
        • Labels
        • Features
        • Examples
      2. Explain how a model is trained and used.
      3. Compare and contrast regression vs. classification.
    3. Problem Framing — This (separate) course reviews some of the material listed above.
      1. Know the types of ML problems and which ones have proven to be particularly difficult.
      2. What things should one think about before trying to frame a machine learning problem (see “Deciding on ML”)?
      Remember that we’ll do the “Try it yourself’ exercises in the lab.

  3. Programming Tools

    1. NumPy — Review the NumPy features:
      1. listed on the Crash Course prerequisites page.
      2. included in the NumPy Tutorial, particularly under sections: “The Basics”; “Shape Manipulation”; “Linear Algebra”.