Safe machine learning model release from Trusted Research Environments: The SACRO-ML package
Smith, Jim, Richard J. Preen, McCarthy, Andrew, Albashir, Maha, Alba Crespi-Boixader, Shahzad Ahmed Mumtaz, Cole, Christian, James Liley, Migenda, Jost, Simon Neave Rogers, Jones, Yola · arXiv (Cornell University) · 2022
We present SACRO-ML, an integrated suite of open source Python tools to facilitate the statistical disclosure control (SDC) of machine learning (ML) models trained on confidential data prior to public release. SACRO-ML combines (i) a SafeModel package that extends commonly used ML models to provide ante-hoc SDC by assessing the vulnerability of disclosure posed by the training regime; and (ii) an Attacks package that provides post-hoc SDC by rigorously assessing the empirical disclosure risk of a model through a variety of simulated attacks after training. The SACRO-ML code and documentation are available under an MIT license at https://github.com/AI-SDC/SACRO-ML