Awesome

Formalizing and Estimating Distribution Inference Risks

This is the code for reproducing the experiments reported in this paper:

Anshuman Suri and David Evans. Formalizing and Estimating Distribution Inference Risks (in Privacy Enhancing Technologies Symposium (PETS) 2022, July 2022.

Exploratory experiments and meta-classifier property-inference attacks on various datasets across several domains.

Folder structure:

• boneage : Experiments on the RSNA Bone-Age task, reduced to binary prediction
• arxiv : Experiments on the ogbn-arxiv node-classfication task
• census : Experiments on the Adult Income dataset
• celeba : Experiments on the Celeb-A dataset
• botnets: Experiments on HarvardNLP's Botnet Detection datasets

Each folder has its own README.md file that explains how to reproduce those experiments.