Federated Learning for Ranking Browser History Suggestions

TL;DR

This paper demonstrates how federated learning can be effectively used to improve browser history suggestion ranking in Firefox while preserving user privacy, leading to more efficient and user-friendly search experiences.

Contribution

The paper introduces a federated learning approach to replace heuristics in Firefox's URL suggestion ranking, ensuring privacy and robustness during deployment.

Findings

Users type over half a character less to find URLs

The system is robust and deployable without degrading user experience

Federated learning effectively trains models for privacy-sensitive browser features

Abstract

Federated Learning is a new subfield of machine learning that allows fitting models without collecting the training data itself. Instead of sharing data, users collaboratively train a model by only sending weight updates to a server. To improve the ranking of suggestions in the Firefox URL bar, we make use of Federated Learning to train a model on user interactions in a privacy-preserving way. This trained model replaces a handcrafted heuristic, and our results show that users now type over half a character less to find what they are looking for. To be able to deploy our system to real users without degrading their experience during training, we design the optimization process to be robust. To this end, we use a variant of Rprop for optimization, and implement additional safeguards. By using a numerical gradient approximation technique, our system is able to optimize anything in Firefox that is currently based on handcrafted heuristics. Our paper shows that Federated Learning can be used successfully to train models in privacy-respecting ways.