Abstract | ||
---|---|---|
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. |
Year | Venue | DocType |
---|---|---|
2019 | arXiv: Computation and Language | Journal |
Volume | Citations | PageRank |
abs/1903.10635 | 1 | 0.35 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mingqing Chen | 1 | 35 | 5.51 |
Rajiv Mathews | 2 | 12 | 5.30 |
Tom Yu Ouyang | 3 | 42 | 2.69 |
Françoise Beaufays | 4 | 341 | 27.76 |