“This is a practical, hands-on workbook filled with math problems that help readers practice the equations behind core deep learning algorithms. It works well as a companion to a traditional textbook (pairing theory with applied exercises) or as a standalone resource for self-study. The problems start with simple examples and gradually increase in complexity, allowing readers to build confidence step by step. I think that this approach is effective for understanding how values are transformed as they pass through different layers of a neural network.

I recommend this book to readers who are new to deep learning, as well as to those who want to refresh their understanding of the mathematical foundations, especially matrix operations and related concepts.”

Andrey Lukyanenko, Kaggle Competition Master, Machine Learning Engineer at Meta

“An exceptional workbook with clean diagrams and color-coded elements that make complex concepts incredibly clear. The brilliant pedagogical progression masterfully builds from dot products to gradients, while hands-on exercises create genuine understanding through active engagement. Highly recommended for anyone looking to deepen their understanding of machine learning fundamentals through hands-on practice.”

Chorouk Malmoum, AI and Data Science Engineer, Founder of AgentX Academy

“Tom Yeh’s Deep Learning Math Workbook is a rare blend of clarity and rigor. It breaks down complex neural network math into tactile, visual exercises that actually teach intuition rather than formula memorization. As someone who’s spent two decades explaining models to quants and traders, I can say this workbook succeeds where most textbooks fail. It makes the math of deep learning feel concrete, approachable, and fun.”

Jason Strimpel , Managing Director, AI & Advanced Analytics, Author of Python for Algorithmic Trading Cookbook

“Prof. Yeh's workbook turns deep learning math into a series of small, satisfying wins.By working through each puzzle step by step, you are not just memorizing formulas, you are internalizing how the pieces fit together, creating a kind of mathematical intuition that stays with you. It’s the hands-on foundation I always wished I had!”

Serg Masis, Principal AI Scientist and Author of Interpretable Machine Learning with Python

“AI by Hand by Prof. Tom Yeh is one of the most intuitive and visual introductions to AI and deep learning you’ll ever find, a workbook that literally lets you “see” how neural networks think. Tom has a unique gift for turning complex math into playful, hands-on puzzles, making the learning experience not only simple but genuinely fun. Anyone who wants to truly understand AI, not just use it, should read this book.”

Eduardo Ordax, Generative AI Lead at AWS and Founder Thinkfluencer AI

“The single best way to learn neural networks (or any topic in computer science) is to build them from scratch. Every resource that aims to help you build a rock-solid foundation is a breath of fresh air in the sea of high-level tutorials, but Prof. Tom Yeh's work stands out from all. If you want to truly master neural networks, the Deep Learning Math Workbook is a must-read.”

Tivadar Danka, Educator in Mathematics & Machine Learning and Author of Mathematics of Machine Learning

“This book feels like opening the hood of deep learning and finally seeing every moving part, dot products, activations, neurons, layers, all laid out with crystal-clear visuals that make the math click instantly. It’s the rare kind of technical book that teaches by revealing, not overwhelming: every diagram builds intuition, every example unlocks another piece of how neural networks actually work.

If you’ve ever wanted a book that turns the machinery of AI into something you can literally follow with your finger, this is it.

A beautifully designed deep-learning guide that makes the math feel alive.”

Mike Erlihson, PhD, Head of AI at Stealth Startup

Tom Yeh is an Associate Professor of Computer Science at the University of Colorado Boulder, specializing in human computer interaction and its societal impacts.