Applied Machine Learning Explainability Techniques

Leverage top XAI frameworks like LIME, SHAP, and others to explain your machine learning models with ease.

Key Features

• Understand the trade-off between model complexity and explainability
• Understand the importance of monitoring production models and checking for Data Drifts
• Use SHAP and LIME framework to make Machine Learning models more interpretable for practical problems

Book Description

As more organizations start adopting AI for their critical business decision-making process, it becomes an immediate expectation to understand and demystify BlackBox AI algorithms. So, learning Explainable AI (XAI) would bring AI algorithms closer to end-users, so that even non-expert users can easily unravel and understand the decision-making process of AI systems.

ML and AI experts working with Data Science, Machine Learning, Deep Learning, and Artificial Intelligence will be able to put their knowledge to work with this practical guide for demystifying the decision-making process of AI. The book provides a hands-on approach to implementation and associated methodologies related to XAI that will have them up-and-running, and productive in no time.

Initially, the readers will understand Explainable AI and its necessity and will get the necessary practical exposure to utilize XAI in the AI/ML problem-solving process by making use of state-of-the-art methods and frameworks. Finally, the readers will get the necessary guidelines to take XAI to the next step by including XAI in the process lifecycle for building any AI/ML system.

By the end of this book, the readers will be able to implement XAI methods and approaches using Python for solving industrial problems, addressing the key pain points encountered, and the best practices in the AI/ML life cycle.

What you will learn

• Learn the principles, factors and various criteria for explainability
• Learn Model Agnostic XAI for structured and unstructured data analysis.
• Hands-on exposure to LIME, SHAP, TCAV, ALIBI, DALEX, InterpretML, ELI5, DiCE
• Learn and make the outcome of AI algorithms explainable to end-users
• Practical problem solving using Explainable AI approaches
• Learn XAI in the development cycle of AI and ML applications and systems

Who This Book Is For

This book is suitable for Data and AI Scientists, AI/ML Engineers with foundational knowledge on Python, Machine Learning, Deep Learning, and Data Science

Table of Contents

1. Foundational Concepts of Explainability Techniques
2. Model Agnostic Explainability
3. Data-Centric Approaches
4. Introduction to LIME for model interpretability
5. Practical exposure of using LIME in ML
6. Introduction to model interpretability using SHAP
7. Practical exposure of using SHAP in ML
8. Human-friendly explanations with TCAV
9. Other popular XAI frameworks
10. XAI Best practices
11. Bridging the AI and end-user gap