What should you consider when building an enterprise federated learning system?

Introduction

Companies like Google and Apple have pioneered federated learning as a way to build higher performing machine learning models on distributed datasets without compromising privacy. Today, Google uses federated learning to power keyboard predictions in Gboard and Apple uses it to improve the accuracy of Face ID and Siri.
But how do you get started?
While there are many great resources that describe what federated learning is, there isn’t a lot of information covering how to apply it in your business.
This article is meant to be a guide that will enable you to set up a scalable federated learning system. Because requirements may differ across users and use cases, this guide won’t provide you with all of the answers. However, it should equip you with key questions and considerations to help you design a system that works for you.
At integrate.ai (where I am Engineering Lead) we are focused on making federated learning more accessible. Here are the seven steps that we’ve uncovered:

Step 1: Pick your model framework

Step 2: Determine the network mechanism

Step 3: Build the centralized service

Step 4: Design the client system

Step 5: Set up the training process

Step 6: Establish the model management system