Skip to content
/ PRICURE Public

PRICURE: Privacy-Preserving Collaborative Inference in a Multi-Party Setting

License

Notifications You must be signed in to change notification settings

um-dsp/PRICURE

Repository files navigation

PRICURE: Privacy-Preserving Collaborative Inference in a Multi-Party Setting

Accompanying code for our paper: Ismat Jarin, Birhanu Eshete. PRICURE: Privacy-Preserving Collaborative Inference in a Multi-Party Setting. In 7th ACM International Workshop on Security and Privacy Analytics (IWSPA'21) co-located with ACM CODASPY'21.

The code in Hospital_Stays_SMPC_DP_upload.ipynb is for the MIMC medical dataset. The same code can be re-purposed for MNIST, Fashin-MNIST, and IDC datasets.

How to reproduce the experiment:

Everything you need is in the notebook: Hospital_Stays_SMPC_DP_upload.ipynb

Contact:

Ismat Jarin: ijarin@umich.edu

Releases

No releases published

Packages

No packages published