This repository contains the code for the paper "Differentially Private Multi-label Learning Is Harder Than You'd Think" by Benjamin Friedl and Anika Hannemann.
Aggregation.py implements the 3 aggregation mechanisms GNThreshold, Confident GNThreshold and Interactive GNThreshold. These are PATE methods generalized for multi-label classification. Privacy.py contains the privacy analysis.
Installation can be done via pip install multilabelpate