I am on my second reading of this insightful and information-packed book by Pedro Domingos.  He masterfully outlines the history of the five “tribes” of Machine Learning (ML), how they have evolved, and where they stand in today’s world. You will gain insight into how all the major players in today’s world, such as Amazon, Google, Facebook, LinkedIn, and many others have evolved, and what to expect in the future.   The quote that hooked me was this one on page 9:

`All of the important ideas in machine learning can be expressed math free. As you read this book, you may even find yourself inventing your own learning algorithms, with nay an equation in sight.”

The Master Algorithm

Tribe #1: Symbolists.  “For Symbolists, all intelligence  can be reduced  to manipulating symbols…Their master algorithm is inverse deduction.”

Tribe #2: Connectionists. ” For Connectionists, learning is what the brain does, so what we need to do is reverse-engineer it.  The Connectionists’ algorithm is “backpropagation”, which compares a system’s output with the desired one  and then successively changes the connections in layer after laye rof neurons so as to bring the output closer to what it should be.”

Tribe #3: Evolutionaries. “Evolutionaries” believe that the mother of all learning is natural selection. If it made us, it can make everything… Their master algorithm is genetic programming, which mates and evolves computer programs in the same way that nature mates and evolves organisms.”

Tribe #4: Bayesians.  “Bayesians are concerned above all with uncertainty.  All learned knowledge is uncertain, and learning itself is a form of uncertain inference…the master algorithm is Bayes’ theorem and its derivates. Bayes’ theorem tells us how to incorporate new evidence into our beliefs, and probabilistic inference algorithms do that as efficiently as possible.”

Tribe #5: Analogizers. ” For Analogizers, the key to learning is recognizing similarities between situations and therbt inferring similarities…The analogizer’s master algorithm is the support vector machine, which figures out which experiences to remember and how to combine them to make new predictions.”

More to come in Part 2, where I discuss how these lessons fit into “Agile Machine Learning.”

-Owen Dall



Posted by Owen Dall

Founder and CTO

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s