Paper restates robust membership inference for mean estimation
The attacks remain effective when released statistics are highly noisy, for Gaussians and binary products.
Stories per day · 30 days
Relateddifferential-privacy1hermite-tensors1learning-theory1mean-estimation1membership-inference1pac-learning1privacy1relu1
3 stories, newest first · one card per story
Stories tagged #gaussian
Fully polynomial algorithm learns positive linear combinations of ReLUs
The method uses poly(d, k, 1/ε) samples and time under Gaussian inputs and L2 loss, without well-conditioned weight assumptions.
Gap-free Differentially Private PCA for Gaussian Data
Researchers present a gap-free differentially private PCA algorithm for Gaussian data.