Building a privacy-preserving Federated Recommender system for mobile devices
For mobile app developers needing privacy-compliant personalization, this work provides a practical, production-ready federated architecture.
The authors propose a two-stage federated recommender system for mobile devices that separates non-sensitive from sensitive data, keeping the latter on-device. On MovieLens, UCI HAR, and a proprietary dataset, they achieve competitive accuracy while preserving privacy.
Serving personalized content on mobile devices has traditionally required pooling sensitive user data on centralized servers, a practice increasingly at odds with modern privacy expectations and geographical regulations. We present a two-stage federated recommendation system pipeline for mobile devices, built around a principled separation between non-sensitive user preference data and sensitive mobile context data that never leaves the device. The first stage runs a collaborative filtering model on non-sensitive app-context data in the cloud to generate a shortlist of relevant items. The second stage re-ranks these candidates on-device using sensitive mobile signals, with only model updates/gradients ever leaving the device. We validate the approach on MovieLens, UCI Human Activity Recognition, and a proprietary pilot dataset, and deliver a production-ready implementation as a Kotlin Multiplatform library deployable on Android and iOS.