Machine Learning changes the way you think about a problem. The focus shifts from a mathematical science to a natural science, running experiments and using statistics, not logic, to analyse its results. — P. Norvig, “Introduction to ML”, Google ML Crash Course.

  1. Google’s Machine Learning Crash Course

    1. Validation Set
      1. Compare and contrast train vs. validation vs. test datasets.
    2. Representation
      1. Terms:
        • Feature vector
        • One-hot vs. multi-hot encodings
        • Binning
      2. What are the qualities of good features?
      3. What are the best practices for data cleansing?
    3. Feature Crosses
      1. Are the logical functions we discussed in class (i.e., AND, OR, XOR) linear functions?
      2. Compare and contrast synthetic features vs. feature crosses.
      3. How are feature crosses useful?

  2. Programming Tools

    1. Keras
      1. Compare and contrast TensorFlow vs. Keras.
      2. What are the guiding principles of Keras?
      3. 30 seconds to Keras — Do these exercises.
    2. Predicting house prices: a regression example — We’ll do this exercise in this unit’s lab. For now, read the exercise and review the following concepts.
      1. The Boston housing dataset
      2. K-fold Validation
      3. MatPlotLib