Federated Learning for Emoji Prediction in a Mobile Keyboard

TL;DR

This paper demonstrates that federated learning enables training of effective emoji prediction models on mobile devices, maintaining user privacy while improving performance over server-trained models.

Contribution

It introduces a federated learning framework for emoji prediction, incorporating transfer learning and mechanisms for triggering and diversifying emoji suggestions.

Findings

Federated learning outperforms server-trained models in emoji prediction accuracy.

Transfer learning improves the model's ability to predict emojis.

The approach maintains user data privacy on mobile devices.

Abstract

We show that a word-level recurrent neural network can predict emoji from text typed on a mobile keyboard. We demonstrate the usefulness of transfer learning for predicting emoji by pretraining the model using a language modeling task. We also propose mechanisms to trigger emoji and tune the diversity of candidates. The model is trained using a distributed on-device learning framework called federated learning. The federated model is shown to achieve better performance than a server-trained model. This work demonstrates the feasibility of using federated learning to train production-quality models for natural language understanding tasks while keeping users' data on their devices.