Information-Driven Machine Learning: Data Science as an Engineering Discipline

This groundbreaking book introduces information measurement methodologies that revolutionize machine learning by reducing conjectures such as magic numbers (hyper-parameters) or model-type bias. It enables data quality measurements, a priori task complexity estimations, and reproducible design of data science experiments, leading to significant size reduction, increased explainability, and enhanced resilience of models. It bridges the gap between machine learning and disciplines such as physics, information theory, and computer engineering, advocating for systematic methodologies grounded in fundamental principles. Ideal for academia and industry professionals, it serves as a valuable tool for those seeking to deepen their understanding of data science as an engineering discipline.

Format: Hardback
Length: 267 pages
Publication date: 02 December 2023
Publisher: Springer International Publishing AG

This groundbreaking book takes machine learning to new heights by introducing innovative information measurement methodologies that revolutionize the field. Stemming from a UC Berkeley seminar on experimental design for machine learning tasks, these techniques aim to break the black box approach of machine learning by reducing conjectures such as magic numbers (hyper-parameters) or model-type bias. Information-based machine learning empowers data quality measurements, a priori task complexity estimations, and reproducible design of data science experiments. The benefits include substantial size reduction, increased explainability, and enhanced resilience of models, all contributing to advancing the discipline's robustness and credibility.

While bridging the gap between machine learning and disciplines such as physics, information theory, and computer engineering, this textbook maintains an accessible and comprehensive style, making complex topics digestible for a broad readership.

Information-Driven Machine Learning delves into the intricate interplay between these disciplines to enhance our understanding of data science modeling. Instead of solely focusing on the how, this text provides answers to the why questions that permeate the field, shedding light on the underlying principles of machine learning processes and their practical implications. By advocating for systematic methodologies grounded in fundamental principles, this book challenges industry practices that have often evolved from ideologic or profit-driven motivations. It addresses a range of topics, including deep learning, data drift, and MLOps, using fundamental principles such as entropy, capacity, and high dimensionality.

Ideal for both academia and industry professionals, this textbook serves as a valuable resource for anyone seeking to deepen their knowledge and expertise in machine learning.

Weight: 606g
Dimension: 235 x 155 (mm)
ISBN-13: 9783031394768
Edition number: 1st ed. 2024