68,885 experimental evaluations of six noise-injection privacy strategies
(CANO, Gaussian, FGSM, PGD, Laplace, Carlini-Wagner) against three adaptive
attacker models (Random Forest, Gradient Boosting, MLP) across 12 datasets,
including the real FP-Stalker browser-fingerprint corpus
(776 users, 13,674 fingerprints, 34 attributes; Vastel et al., IEEE S&P 2018).
Aggregate statistics in the paper are computed over 54,281 in-scope
configurations after… See the full description on the dataset page:
https://huggingface.co/datasets/datamatters24/cano-adversarial-privacy.
