Linear Algebra and Learning from Data

by Gilbert Strang

Linear algebra and the foundations of deep learning, together at last! From Professor Gilbert Strang, acclaimed author of Introduction to Linear Algebra, comes Linear Algebra and Learning from Data, the first textbook that teaches linear algebra together with deep learning and neural nets. This readable yet rigorous textbook contains a complete course in the linear algebra and related mathematics that students need to know to get to grips with learning from data. Included are: the four fundamental subspaces, singular value decompositions, special matrices, large matrix computation techniques, compressed sensing, probability and statistics, optimization, the architecture of neural nets, stochastic gradient descent and backpropagation.

Genres: Mathematics, Textbooks, Artificial Intelligence, Algebra, Science, Computer Science, Nonfiction

446 pages, Hardcover

Published January 31, 2019

About the author

William Gilbert Strang (born November 27, 1934), usually known as simply Gilbert Strang or Gil Strang, is an American mathematician, with contributions to finite element theory, the calculus of variations, wavelet analysis and linear algebra. He has made many contributions to mathematics education, including publishing seven mathematics textbooks and one monograph. Strang is the MathWorks Professor of Mathematics at the Massachusetts Institute of Technology. He teaches Introduction to Linear Algebra and Computational Science and Engineering and his lectures are freely available through MIT OpenCourseWare.

Reviews

[Saurabh Jha · Rating 1 out of 5]

I had a background in undergrad Linear Algebra from one of the Strang's book "Introduction to Linear Algebra" before reading this book. Even with that background, I haven't been able to make sense of it, even after watching the accompanying lectures.

The basic problem is Strang loves to introduce an idea without explaining it, perhaps for motivation or to give readers the sneak peek to readers of the wonderful things that await them. In this book, however, he has taken it too far. It's just one idea after another without any explanation, why is the idea useful, and where is it coming from? In the end, I think the struggle to make sense of this book is not worth it.

[Sean Richardson · Rating 5 out of 5]

Used with the lectures as a linear algebra refresher.

[Corbin Smith · Rating 3 out of 5]

Its not finished

[estrabismo · Rating 5 out of 5]

This is probably my current favorite textbook as far as applications. I use it as a reference all the time. It is for an audience who as already gone through a linear algebra course/ is familiar with key topics. It's not meant to be exhaustive. It's always so enjoyable/fun to read Gil Strang because his excitement about linear algebra is so present. That's what also makes his writing so effective/engaging, he really sees a lot of beauty in it and aims to get you to see it too. I find all of the parts very current, salient, and essential for entering data science or thinking about data.