Gap-free Differentially Private PCA for Gaussian Data

תקציר מקורי באנגליתarXiv:2609.31614v2 Announce Type: replace-cross Abstract: We give a gap-free $(\epsilon,\delta)$-differentially private algorithm for the principal component analysis (PCA) problem with Gaussian data. The algorithm is based on a private variant of the power iteration method, and it is computationally efficient.