Machine learning is transforming fields from healthcare diagnostics to climate change predictions through their predictive performance. However, these complex machine learning models often lack interpretability, which is becoming more essential than ever for debugging, fostering trust, and communicating model insights.

Introducing SHAP, the Swiss army knife of machine learning interpretability:

• SHAP can be used to explain individual predictions.
• By combining explanations for individual predictions, SHAP allows to study the overall model behavior.
• SHAP is model-agnostic – it works with any model, from simple linear regression to deep learning.
• With its flexibility, SHAP can handle various data formats, whether it’s tabular, image, or text.
• The Python package shap makes the application of SHAP for model interpretation easy.

This book will be your comprehensive guide to mastering the theory and application of SHAP. It starts with the quite fascinating origins in game theory and explores what splitting taxi costs has to do with explaining machine learning predictions. Starting with using SHAP to explain a simple linear regression model, the book progressively introduces SHAP for more complex models. You’ll learn the ins and outs of the most popular explainable AI method and how to apply it using the shap package.

In a world where interpretability is key, this book is your roadmap to mastering SHAP. For machine learning models that are not only accurate but also interpretable.

Who This Book Is For

This book is for data scientists, statisticians, machine learners, and anyone who wants to learn how to make machine learning models more interpretable. Ideally, you are already familiar with machine learning to get the most out of this book. And you should know your way around Python to follow the code examples.

What's in the Book

Note: Please be aware that the ePub version utilizes MathML for mathematical notations and may not be compatible with all eReaders. Leanpub has a 60-day "100% Happiness Guarantee", so don't hesitate to just try it out. And you'll also get the PDF where the equations look good.

1. Introduction
2. A Short History of Shapley Values and SHAP
3. Theory of Shapley Values
4. From Shapley Values to SHAP
5. Estimating SHAP Values
6. SHAP for Linear Models
7. Classification with Logistic Regression
8. SHAP for Additive Models
9. Understanding Feature Interactions with SHAP
10. The Correlation Problem
11. Regressing Using a Random Forest
12. Image Classification with Partition Explainer
13. Image Classification with Deep and Gradient Explainer
14. Explaining Language Models
15. Limitations of SHAP
16. Building SHAP Dashboards with Shapash
17. Alternatives to the shap Library
18. Extensions of SHAP
19. Other Applications of Shapley Values in Machine Learning
20. SHAP Estimators
21. The Role of Maskers and Background Data

About me (Christoph Molnar)

Author of the free online book Interpretable Machine Learning. I have a background in both statistics and machine learning and did my Ph.D. in interpretable machine learning. After a mix of data scientist jobs and academia, I'm now a full-time machine learning book author.

Interpretable Machine Learning is a comprehensive guide to making machine learning models interpretable

"Pretty convinced this is the best book out there on the subject"

– Brian Lewis, Data Scientist at Cornerstone Research

Summary

This book covers a range of interpretability methods, from inherently interpretable models to methods that can make any model interpretable, such as SHAP, LIME, and permutation feature importance. It also includes interpretation methods specific to deep neural networks and discusses why interpretability is important in machine learning. All interpretation methods are explained in depth and discussed critically. How do they work under the hood? What are their strengths and weaknesses? How can their outputs be interpreted?

"What I love about this book is that it starts with the big picture instead of diving immediately into the nitty gritty of the methods (although all of that is there, too)."

– Andrea Farnham, Researcher at Swiss Tropical and Public Health Institute

Who the book is for

This book is essential for machine learning practitioners, data scientists, statisticians, and anyone interested in making their machine learning models interpretable. It will help readers select and apply the appropriate interpretation method for their specific project.

"This one has been a life saver for me to interpret models. ALE plots are just too good!"

– Sai Teja Pasul, Data Scientist at Kohl's

You'll learn about

• The concepts of machine leaning interpretability
• Inherently interpretable models
• Methods to make any machine model interpretable, such as SHAP, LIME and permutation feature importance
• Interpretation methods specific to deep neural networks
• Why interpretability is important and what's behind this concept

About the author

The author, Christoph Molnar, is an expert in machine learning and statistics, with a Ph.D. in interpretable machine learning.

Other Versions

The print version can be bought on Amazon.

A free HTML version of the book can be found at: https://christophm.github.io/interpretable-ml-book/