Название: Tuning Up: From A/B testing to Bayesian optimization (MEAP) Автор: Dawid Sweet Издательство: Manning Publications Год: 2020 Формат: PDF, MOBI Страниц: 107 Размер: 10 Mb Язык: English
Master industry-proven tests, methods, and evaluative experiments to deliver continuous improvements to your software. In Tuning Up: From A/B testing to Bayesian optimization you will learn how to: Design, run, and analyze an A/B test Assess the effectiveness of a new feature Increase experimentation rate with multi-armed bandits Tune multiple parameters experimentally with Bayesian optimization Clearly define business metrics used for decision making Identify and avoid the common pitfalls of experimentationv
Tuning Up: From A/B testing to Bayesian optimization is a toolbox of experimental methods that will keep your software and systems working at peak performance. You’ll learn to implement tests and techniques that will boost the effectiveness of machine learning systems, trading strategies, infrastructure, and more. Each method in this practical guide is regularly utilized in highly competitive industries like finance and social media. About the Technology Tuning your software and systems is best done by following established methods employed by high-performing teams like the ones led by author David Sweet. This book reveals tests, metrics, and practical tools that will ensure your projects are constantly improving, delivering revenue, and ensuring user satisfaction. About the book Tuning Up: From A/B testing to Bayesian optimization teaches you proven methods for improving your software and data systems. Each method has been tested in industry, and is fully explained with easy-to-understand math and Python code—no black boxes you just have to trust are working! The book is filled with real-world use cases for quantitative trading, recommender systems, and social media. You’ll learn how to evaluate changes to your system and explore ways to ensure that your testing is not undermining revenue and other business metrics. By the time you’re done, you’ll be able to seamlessly run effective performance experiments whilst avoiding common mistakes and pitfalls.