Learning Differentially Private Recurrent Language Models

TLDR

Recent advances in training deep networks on user‑partitioned data and privacy accounting for stochastic gradient descent underpin this work. The authors add user‑level privacy to federated averaging, enabling large‑step updates from user data. The study shows that large recurrent language models can be trained with user‑level differential privacy at negligible accuracy loss, that privacy mainly increases computation rather than reducing utility, and that private LSTM models are comparable to un‑noised models on large datasets.

Abstract

We demonstrate that it is possible to train large recurrent language models with user-level differential privacy guarantees with only a negligible cost in predictive accuracy. Our work builds on recent advances in the training of deep networks on user-partitioned data and privacy accounting for stochastic gradient descent. In particular, we add user-level privacy protection to the federated averaging algorithm, which makes "large step" updates from user-level data. Our work demonstrates that given a dataset with a sufficiently large number of users (a requirement easily met by even small internet-scale datasets), achieving differential privacy comes at the cost of increased computation, rather than in decreased utility as in most prior work. We find that our private LSTM language models are quantitatively and qualitatively similar to un-noised models when trained on a large dataset.