Software

I develop and maintain open-source tools that make privacy-preserving methods and machine learning techniques accessible to researchers. All code is available on GitHub.

RGAN: Generative Adversarial Networks in R

R package for training Generative Adversarial Networks, available on CRAN with 8,300+ downloads.

MIBench: Multiple Imputation Benchmark

First standardized benchmark for evaluating multiple imputation algorithms across diverse scenarios.

Private Post-GAN Boosting

Replication code for ICLR 2021 paper on differentially private synthetic data generation.