Hands-On Federated Analytics

A practical guide to secure collaborative data science

Welcome

It is a truth universally acknowledged, that a data scientist in possession of good skills, must be in want of more data.

Why I wrote this book:

• Federated analytics large potential but notoriously hard and often misunderstood. ‘Federation’ can mean many things.
• Motivated to support data science for the common good, I believe that federated analytics has a role to play in implementing data cooperatives, enable more wide-scale data sharing to support transformative change in healthcare, education, agriculture etc.
• Interesting from teaching modern data science: it touches on many separate topics that are interesting but hard to see in context. Cryptography, cloud engineering, standardization, predicate pushdown…
• Aim to enlarge the frame of thinking of those interested in the field. Side note denkraam with comic Maarten Toonder.

Scope: federated analytics at large. Having better understating of how federated queries work is also a huge step forward towards data commons.

Structure:

• stand-alone chapters that can be read on their own.
• chapter 1 to 4 explain basic principles and concepts of federated learning
• chapter 5 to 8 are specific usecases/applications
• chapter 9 and 10 are more technical and aimed at engineers and system administrators how to implement and maintain FA
• chapter 11 and 12 address ongoing and future developments
• accompanying code respository to run various examples and use-case
  • using vantage6 as the main platform

Cover image credit

In the spirit of the animal menagerie of O’Reilly books, I have chosen the Fortuna Fragilis as the preliminary cover. This species can be found on Terra Ultima by Raoul Deleo. I like to think that collaborative data science is fragile and fortuitous at the same time, and that it is worth kindling if we are to put data to use for the common good.