Learning Differentially Private Recurrent Language Models

H. Brendan McMahan, Daniel Ramage, Kunal Talwar, Li Zhang

Feb 15, 2018 (modified: Feb 24, 2018) ICLR 2018 Conference Blind Submission readers: everyone Show Bibtex
  • 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.
  • TL;DR: User-level differential privacy for recurrent neural network language models is possible with a sufficiently large dataset.
  • Keywords: differential privacy, LSTMs, language models, privacy