chapter one

1 Why causal AI

 

This chapter covers

  • Defining causal AI and its benefits
  • Incorporating causality into machine-learning models
  • A simple example of applying causality to a machine-learning model

Subscription streaming platforms like Netflix are always looking for ways to optimize various indicators of performance. One of these is their churn rate, meaning the rate at which they lose subscribers. Imagine you were a machine learning engineer or data scientist at Netflix tasked with finding ways of reducing churn. What are the causal questions (questions that require causal thinking) you might ask with respect to this task? Some of these questions might include:

  • Causal discovery. Given detailed data on who churned and who did not, can you analyze that data to find causes of churn? Causal discovery is the question of what causes what.
  • Estimating average treatment effects. Suppose the choice of algorithm that recommends content to the user is a cause of the churn rate; a better choice of algorithm reduces churn. But by how much on average? The task of quantifying how much on average a cause drives an effect is average treatment effect estimation. For example, some users are exposed to a new version of the algorithm. How much does this impact churn relative to a baseline algorithm?

1.1 What is causal AI?

1.2 How this book approaches causal inference

1.2.1 Target audience

1.2.2 Overview of the book

1.2.3 Distinctive features of this book

1.3 Causality’s role in modern AI workflows

1.3.1 Better data science

1.3.2 More robust, decomposable, and explainable models

1.3.3 Fairer AI

1.4 How causality is driving the next AI wave

1.5 A machine learning-themed primer on causality

1.5.1 Queries, probabilities, and statistics

1.5.2 Causality and MNIST

1.5.3 Causal queries, probabilities, and statistics

1.6 Summary