Cover of 'Deep Learning and the Game of Go' and its Japanese translation

I wrote a book! (Please buy it)

I’ve been interested in game AI ever since I was a kid, and I’ve played go off-and-on since I was a teenager. So when the AlphaGo news hit, I was obsessed; I stayed up late to watch the games against Lee Sedol, scrutinized the game records, and pored over all the papers. It was my pal Max’s idea to turn our hobby projects into a book.

Go? What?

Go is the best game ever invented. It’s a pure abstract strategy game (like chess, but better). It originated in China thousands of years ago. It’s widely played throughout East Asia today, with a small but dedicated following elsewhere around the world.

Go was a notoriously difficult game for computers, but in 2016, the AlphaGo AI beat one of the all-time greatest players with a clever combination of deep learning and traditional tree search.

I can’t believe this got published

OK here is my pitch for the book if you are not a giant go nerd.

There are lots of books about machine learning, but they mainly take you through toy examples. This book really focuses on the practical aspects of deep learning:

  • Finding training data
  • Turning wild data into a machine-readable format
  • Tuning network architecture
  • Feature engineering
  • Integrating deep learning with a traditional algorithm
  • Building a useful app around a deep learning core

Even if you’re not a go player, my claim is that these are under-covered concepts that apply to all kinds of machine learning projects.