Interpretable Machine Learning - A Guide for Making Black Box Models Explainable

Sep 4, 2026

Uploaded by Eden

Interpretable Machine Learning explains how to transform opaque machine learning predictions into models whose decisions can be understood. The guide covers both simple interpretable methods, including decision trees, decision rules, and linear regression, and general model-agnostic approaches for black boxes such as feature importance and accumulated local effects. It also details local explanation techniques like Shapley values and LIME, with an emphasis on strengths, weaknesses, evaluation, and correct interpretation for tabular data.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages251

Reading Time10 h 33 min

AccessFree to download

Preface

Introduction

Story Time

What Is Machine Learning?

Interpretability

Taxonomy of Interpretability Methods

Evaluation of Interpretability

Interpretable Models

Linear Regression

Logistic Regression

Decision Tree

RuleFit

Model-Agnostic Methods

Partial Dependence Plot (PDP)

Individual Conditional Expectation (ICE)

Accumulated Local Effects (ALE) Plot

Feature Interaction

Feature Importance

Global Surrogate

Local Surrogate (LIME)

Shapley Values

Example-Based Explanations

Counterfactual Explanations

Prototypes and Criticisms

The Future of Machine Learning

Interpretable Machine Learning A Guide for Making Black Box Models Explainable Christoph Molnar This book is for sale at [http://leanpub.com/interpretable-machine-learning](http://leanpub.com/interpretable-machine-learning)[ ](http://leanpub.com/interpretable-machine-learning)[This version was published on 2019-02-21](This version was published on 2019-02-21) This is a Leanpub book. Leanpub empowers authors and publishers with the Lean Publishing process. Lean Publishing is the act of publishing an in-progress ebook using lightweight tools and many iterations to get reader feedback, pivot until you have the right book and build traction once you do. This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4 .0 International License Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Introduction .................................................. 2 Story Time ................................................. 4 What Is Machine Learning? ...................................... 10 Terminology ................................................ 12 Interpretability ................................................ 15 Importance of Interpretability ..................................... 15 Taxonomy of Interpretability Methods ................................ 21 Scope of Interpretability ......................................... 23 Evaluation of Interpretability ..................................... 25 Properties of Explanations ....................................... 26 Human-friendly Explanations ..................................... 29 Datasets ..................................................... 34 Bike Rentals (Regression) ........................................ 34 YouTube Spam Comments (Text Classification) .......................... 35 Risk Factors for Cervical Cancer (Classification) .......................... 36 Interpretable Models ............................................. 37 Linear Regression ............................................. 38 Logistic Regression ............................................ 54 GLM, GAM and more .......................................... 61 Decision Tree ............................................... 79 Decision Rules .............................................. 85 RuleFit ................................................... 100 Other Interpretable Models ....................................... 108 Model-Agnostic Methods .......................................... 110 Partial Dependence Plot (PDP) ..................................... 113 Individual Conditional Expectation (ICE) .............................. 119 Accumulated Local Effects (ALE) Plot ................................ 125 Feature Interaction ............................................ 146 Feature Importance ............................................ 154 CONTENTS Global Surrogate ............................................. 163 Local Surrogate (LIME) ......................................... 168 Shapley Values .............................................. 177 Example-Based Explanations ....................................... 189 Counterfactual Explanations ...................................... 191 Adversarial Examples .......................................... 199 Prototypes and Criticisms ........................................ 208 Influential Instances ........................................... 218 A Look into the Crystal Ball ........................................ 234 The Future of Machine Learning ................................... 235 The Future of Interpretability ..................................... 237 Contribute to the Book ........................................... 240 Citing this Book ................................................ 241 Acknowledgements ............................................. 242 Refe

FAQs

What is the main goal of this book?

The book focuses on making machine learning models and their predictions interpretable, overcoming the barrier that black box models usually do not explain their outputs.

Which kinds of interpretability methods are covered?

It covers simple interpretable models (e.g., decision trees, decision rules, linear regression) and model-agnostic techniques (e.g., feature importance and accumulated local effects), plus local explanation methods such as Shapley values and LIME.

What data types does the book primarily target?

The book primarily focuses on machine learning models for tabular data (structured/relational data), with less emphasis on computer vision and natural language processing tasks.

Sep 4, 2026

Uploaded by Eden

Interpretable Machine Learning - A Guide for Making Black Box Models Explainable_Page 1

Open in WPS to view all 251 pages