
Robust Explainable AI
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The area of Explainable Artificial Intelligence (XAI) is concerned with providing methods and tools to improve the interpretability of black-box learning models. While several approaches exist to generate explanations, they are often lacking robustness, e.g., they may produce completely different explanations for similar events. This phenomenon has troubling implications, as lack of robustness indicates that explanations are not capturing the underlying decision-making process of a model and thus cannot be trusted.
This book aims at introducing Robust Explainable AI, a rapidly growing field whose focus is to ensure that explanations for machine learning models adhere to the highest robustness standards. We will introduce the most important concepts, methodologies, and results in the field, with a particular focus on techniques developed for feature attribution methods and counterfactual explanations for deep neural networks.
As prerequisites, a certain familiarity with neural networks and approaches within XAI is desirable but not mandatory. The book is designed to be self-contained, and relevant concepts will be introduced when needed, together with examples to ensure a successful learning experience.
More details
Other editions
Additional editions

Persons
Francesco Leofante is a researcher affiliated with the Centre for Explainable AI at Imperial College. His research focuses on explainable AI, with special emphasis on counterfactual explanations for AI-based decision-making. His recent work highlighted several vulnerabilities of counterfactual explanations and proposed innovative solutions to improve their robustness.
Matthew Wicker is an Assistant Professor (Lecturer) at Imperial College London and a Research Associate at The Alan Turing Institute. He works on formal verification of trustworthy machine learning properties with collaborators form academia and industry. His work focuses on provable guarantees for diverse notions of trustworthiness for machine learning models in order to enable responsible deployment.
Content
Foreword.- Preface.- Acknowledgements.- 1. Introduction.- 2. Explainability in Machine Learning: Preliminaries & Overview.- 3. Robustness of Counterfactual Explanations.- 4. Robustness of Saliency-Based Explanations.
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.