Machine Learning: The Basics

Machine learning (ML) is a standard tool for many fields of science and engineering, and this book approaches it as the computational implementation of the scientific principle of continuously adapting a model of a given data-generating phenomenon by minimizing some form of loss incurred by its predictions. Its three-component approach provides uniform coverage of a wide range of concepts and techniques, such as regularization, privacy-preservation, and explainability.

Format: Paperback / softback
Length: 212 pages
Publication date: 23 January 2023
Publisher: Springer Verlag, Singapore

Machine learning (ML) has permeated our daily lives and emerged as a indispensable tool across diverse scientific and engineering disciplines. To maximize the potential of ML, it is crucial to grasp its fundamental principles. This book presents ML as the computational realization of a scientific principle, which involves the iterative refinement of a model to minimize a specific type of loss incurred by its predictions. By adopting a three-component framework, the book equips readers with the ability to dissect diverse ML applications and methodologies in terms of data, model, and loss. This approach facilitates informed decision-making in selecting from the vast array of readily available ML methods.

Within this comprehensive framework, the book delves into various techniques, including regularization, privacy preservation, and explainability, which represent specific design choices for the model, data, and loss of a ML method. These techniques are essential in addressing challenges such as model overfitting, ensuring data confidentiality, and providing interpretability of ML models to stakeholders.

By adopting a rigorous and systematic approach, this book aims to empower readers with a deep understanding of ML principles and their practical applications. Whether you are a researcher, practitioner, or enthusiast seeking to leverage the power of ML, this book serves as a valuable resource for advancing your knowledge and skills in this rapidly evolving field.

Weight: 361g
Dimension: 235 x 155 (mm)
ISBN-13: 9789811681950
Edition number: 1st ed. 2022