Federated Learning Of Out-Of-Vocabulary Words

• Main Authors: Chen, Mingqing, Mathews, Rajiv, Ouyang, Tom, Beaufays, Françoise
• Format: Journal Article
• Language: English
• Published: 25.03.2019
• Subjects: Computer Science - Computation and Language
• DOI: 10.48550/arxiv.1903.10635

We demonstrate that a character-level recurrent neural network is able to learn out-of-vocabulary (OOV) words under federated learning settings, for the purpose of expanding the vocabulary of a virtual keyboard for smartphones without exporting sensitive text to servers. High-frequency words can be sampled from the trained generative model by drawing from the joint posterior directly. We study the feasibility of the approach in two settings: (1) using simulated federated learning on a publicly available non-IID per-user dataset from a popular social networking website, (2) using federated learning on data hosted on user mobile devices. The model achieves good recall and precision compared to ground-truth OOV words in setting (1). With (2) we demonstrate the practicality of this approach by showing that we can learn meaningful OOV words with good character-level prediction accuracy and cross entropy loss.