High-Dimensional Probability

An Introduction with Applications in Data Science

2019 Prose Award for Mathematics

Who is this book for?

This textbook is aimed at doctoral students, advanced master's students, and beginning researchers in mathematics, statistics, computer science, electrical engineering, and related fields, who seek to deepen their understanding of probabilistic methods commonly used in modern data science research. It can be used for self-study or as a textbook for a second probability course with data science applications.

Why this book?

Data science is evolving rapidly, and probabilistic methods are key to these advances. A typical graduate probability course no longer provides the mathematical sophistication needed for early-career data science researchers. This book aims to fill that gap, presenting essential probabilistic methods and results for mathematical data scientists.

Are you ready?

To read this book, you will need a solid knowledge of probability theory (at the masters or doctoral level), strong undergraduate linear algebra, and some familiarity with metric, normed, and Hilbert spaces. Measure theory is not required.

Roman Vershynin

Roman Vershynin

I am Professor of Mathematics at the University of California, Irvine. My research spans high-dimensional probability and mathematical data science.

Take a look at my webpage to learn more. E-mail me.

NEW! The draft of the second edition is now online:

Download the Second Edition

Want to get notified when the printed version is out? Email me: rvershyn@uci.edu

The first eidition in printed form can be purchased on Amazon and in Cambridge University Press. It is freely available online:

Download the First Edition

News