Disc 1. Telling the computer what we want --
Starting with Python Notebooks and Colab --
Decision trees for logical rules --
Neural networks for perceptual rules --
Opening the Black box of neural network --
Bayesian models for probability prediction --
Genetic algorithms for evolved rules --
Disc 2. Nearest neighbors for using similarity --
The fundamental pitfall of overfitting --
Pitfalls in applying machine learning --
Clustering and semi-supervised learning --
Recommendations with three types of learning --
Games with reinforcement learning --
Disc 3. Deep learning for computer vision --
Getting a deep learner back on track --
Text categorization with words as vectors --
Deep networks that output language --
Making stylistic images with deep networks --
Making photorealistic images with GANs --
Disc 3. Deep learning for speech recognition --
Inverse reinforcement learning from people --
Casual inference comes to machine learning --
The unexpected power of over-parameterization --
Protecting privacy within machine learning --
Mastering the machine learning process.
