Better understanding with Optimization for Machine Learning

Long awaited book from Machine Learning Mastery

Recently, I’ve been reading the new Optimization for Machine Learning book from the Machine Learning Mastery written by Jason Brownlee. It just so happened that I read it fully from start to end, since I was one of the technical reviewers of the book. The book was interesting to read thanks to a number of ingredients.

As always Jason was able to write an engaging book with practical advice that can be actioned right away using open source software on Linux, Windows or MacOS. Apart from this the book has just enough clearly explained theoretical material so that even beginning machine learning practitioners can play with optimization algorithms described in the book.

What I liked and what surprised me in the book

Personally, I think it was while reading and working with this book, that I truly understood what an optimizations is. How it is used in machine learning. What is an optimization algorithm, like gradient descent and how to implement one from scratch. I also very much enjoyed the chapter about Global Optimization with various types of Evolution Algorithms. What was funny, that about two weeks after I finished reading the book I came across Donald Hoffman’s The Interface Theory of Perception with relation to consciousness which is based on The Theory of Evolution by Natural Selection. For example, one of his papers written with colleagues namely Does evolution favor true perception? provides an example of Genetic Algorithm (GA) which very much resembles the GA in Chapter 17 of the book. It is highly recommended reading for anyone interested in how consciousness arises in the mind. By the way, does it?

Overall

The Optimization for Machine Learning book is what you come to expect from Machine Learning Mastery books. It’s interesting, it’s practical and it makes you understand what is that you are doing in Machine Learning. As always, each chapter has extensive references to tutorials, technical papers and books on machine learning. So don’t wait and start reading it, maybe you’ll come up with a new theory of how consciousness emerges in the mind.

References

Donald D. Hoffman, Manish Singh, Justin Mark, “Does evolution favor true perceptions?”, Proceedings Volume 8651, Human Vision and Electronic Imaging XVIII; 865104 (2013)