Experimentation for Engineers: From A/B testing to Bayesian optimization

I’ve read quite a few books on the subject that stop at the equivalent of Chapter 3. David Sweet doesn’t disappoint as he dives deeper into advanced topics such as Response Surface Methodology, Contextual Bandits, and Bayesian Optimization. Formulas (in moderation), Python and nice plots, make this book an easy read, that reminds me "Bayesian Methods for Hackers". It doesn’t promise so, but this book might not be complete-beginner-friendly. A practitioner will have to do some intelligent work to get the concepts and apply them to their field.

I must say it is a great book for anyone looking to optimize systems through experimentation. It offers practical experiments used by engineers in the world’s most competitive industries and gives an in-depth look at methods such as A/B testing, multi-armed bandits, response surface methodology, contextual bandits, and Bayesian optimization.It is a great resource for any data scientist or data engineer looking to get the most out of their system, even for traditionally non-technical profiles looking to diversify their knowledge as it is an easy-to-understand lecture and provides clear examples and explanations.. Highly recommend!

This is a practical but rigorous book on creating experiments that drive business outcomes with A/B testing and beyond: multi-armed bandits, contextual bandits, and Bayesian optimization. David Sweet has taught material from this book to M.S. students in both Artificial Intelligence and in Data Analytics and Visualization at my university, with consistently high student feedback. The emphasis on computational and statistical methods for experimentation fills a needed gap in the literature, since most of the previous material on this subject has been written for business analysts, not data scientists. A motivated data scientist or data engineer with basic Python and statistical skills will get a lot of benefit from this book. I highly recommend.