This article was adapted from episode #470 of the SuperDataScience Podcast.
At the end of every guest episode of the SuperDataScience show — so every odd-numbered episode — I ask the guest for a book recommendation. About half of the time, these are related to data science in some way; the other half of the time, they’re about completely unrelated but interesting topics, such as champagne, the hallucinogen cocktails dreamt up by medieval witches, and the biographies of great entrepreneurs.
Given that SuperDataScience listeners love this question, I thought I’d take a few minutes to tell you about some of my own all-time favorite books.
As detailed in Episode #442 of the SuperDataScience Podcast, I’m a big fan of James Clear’s writing, particularly his #1 New York Times bestseller Atomic Habits. You can head back to episode #442 for a longer treatise on the Atomic Habits approach, but in a nutshell: The most efficient and effective way to make big, desired changes across all aspects of your life is by committing to small, measurable adjustments. In that episode, I detail examples of this with respect to data science skills.
While I love Atomic Habits and I recommend it a lot given it’s impact on my own life, I don’t think I could say it’s my absolutely favorite non-fiction work of all time. That distinction is reserved for Yuval Noah Harari’s Sapiens.
Harari delves into numerous topics— anthropological, historical, scientific and philosophical, with occasional emphases on Buddhism and mindfulness — that I reflect on day in and day out. His writing is clear, concise and playful, and the book left me feeling centred and optimistic about projections involving our species (war, automation, biotechnological augmentation) and planet (climate change, ecological destruction, gene editing) that I’ve found troubling for years. For a time, I couldn’t shut up about the facts and ideas I distilled from the text!
Recently, I picked up the first volume of the graphic novel (i.e., comic book-style) version of Sapiens, which was released in late 2020, with other volumes to follow. I haven’t had a chance to read it yet, but expect that it’s awesome, particularly if you enjoy learning visually.
Speaking of which, if you’re into learning visually, I can’t help but mention my book Deep Learning Illustrated, which makes use of over a hundred hand-drawn illustrations, full-color mathematical equations, and hundreds of hands-on code demos to make learning the theory of artificial neural networks as easy and fun as possible.
I’m not here to plug my own book, it was simply a less-than-subtle segue to Goodfellow, Bengio, and Courville’s Deep Learning, which doesn’t have any code examples, but rather is the bible for immersing yourself in the mathematics of deep learning.
The authors have made the book available freely online, but because this math is inaccessible to many folks without an undergraduate degree in the field, I’ve been working away on my Machine Learning Foundations series of lectures and video tutorials, which details the linear algebra, calculus, probability theory, and computer science you need to know to tackle a book like Deep Learning -- or any graduate-level machine learning textbook or academic paper. The series is available for free on YouTube, or — if you’d like to get detailed solution walkthroughs and other goodies like a certificate of completion — you can purchase the typically quite inexpensive Udemy version of the content, which I am publishing in partnership with SuperDataScience.
Finally, I enjoy getting absorbed in works of fiction as well. My all-time favorite is Cat’s Cradle by the satirical American author Kurt Vonnegut.
Vonnegut’s books are typically found in the science fiction section of bookstores, but I think this is a misclassification. He invokes science-fiction concepts only to convey deep philosophical insights and he manages to do so with buckets full of dry humor; I love both of these aspects of his writing. While Vonnegut has many marvellous books, I’d say start with Cat’s Cradle, as that particular book contains philosophies that have shaped my life time and time again.
You can head to jonkrohn.com/bookreviews for many more three-sentence reviews of books I’ve read in recent years.