Trustworthy Machine Learning under Imperfect Data

The subject of this book centres

around trustworthy machine learning under imperfect data. It is primarily designed for

scientists, researchers, practitioners, professionals, postgraduates and

undergraduates in the

field of machine learning and artificial intelligence. The book focuses

on trustworthy deep learning under various types of imperfect data, including

noisy labels, adversarial examples, and out-of-distribution data. It covers

trustworthy machine learning algorithms, theories, and systems.

The main goal of the book is to provide students and researchers in academia with an

unbiased and comprehensive literature review. More importantly, it aims to stimulate

insightful discussions about the future of trustworthy machine learning. By engaging the audience

in more in-depth conversations, the book intends to spark ideas for addressing core

problems in this topic. For example, it will explore how to build up benchmark datasets in

noisy-supervised learning, how to tackle the emerging adversarial learning, and

how to tackle out-of-distribution detection.

For practitioners in the industry,

this book will present state-of-the-art trustworthy machine learning methods to

help them solve real-world problems in different scenarios, such as online

recommendation and web search. While the book will introduce the basics of

knowledge required, readers will benefit from having some familiarity with

linear algebra, probability, machine learning, and artificial intelligence. The

emphasis will be on conveying the intuition behind all formal concepts,

theories, and methodologies, ensuring the book remains self-contained at a high

level.