Code for the ICLR'23 paper: Robust Fair Clustering: A Novel Fairness Attack and Defense Framework

Requirements:

python-mnist
gdown
kmedoids
pulp
torch
scikit-learn==0.22.2
zoopt
pyckmeans

Instructions:

• For the attack section of the paper, please follow the code in Attack.ipynb.
• For the defense section of the paper, please follow the code in Defense.ipynb.
• Using the code provided in these notebooks, the experimental results in the paper can be obtained.
• If there any questions, please reach out to {last_name}@ucdavis.edu where {last_name} is the first author's last name.