The recently published book Understanding Deep Learning by [Simon J. D. Prince] is notable not only for focusing primarily on the concepts behind Deep Learning — which should make it highly accessible to most — but also in that it can be either purchased as a hardcover from MIT Press or downloaded for free from the Understanding Deep Learning website. If you intend to use it for coursework, a separate instructor answer booklet and other resources can be purchased, but student resources like Python notebooks are also freely available. In the book’s preface, the author invites readers to send feedback whenever they find an issue.
In the preface, the joke contained in the title is also explained, as nobody really ‘understands’ deep learning algorithms, even though they are beginning to underlie more and more of modern society, ranging from convolutional neural networks (CNNs) in machine vision to recurrent neural networks (RNNs) commonly used in large language models that often are ascribed near-magical properties which they most definitely do not possess. We’ve likened it to a bad summer intern, although they are getting better almost daily.
The book and materials look like a solid introduction to learning about the ideas behind these deep learning algorithms for anyone even mildly curious about the topic. You can’t complain about the price
Support the originator by clicking the read the rest link below.