Kunal Talwar

Nearly-optimal private LASSO

2015 · Neural Information Processing Systems

We present a nearly optimal differentially private version of the well known LASSO estimator. Our algorithm provides privacy protection with respect to each training example. The excess risk of our algorithm, compared to the non-private …

Learning Differentially Private Recurrent Language Models

2017 · arXiv (Cornell University)

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 …

Encode, Shuffle, Analyze Privacy Revisited: Formalizations and Empirical Evaluation

2020 · arXiv (Cornell University)

Recently, a number of approaches and techniques have been introduced for reporting software statistics with strong privacy guarantees. These range from abstract algorithms to comprehensive systems with varying assumptions and built upon local differential privacy …