pith. sign in

By Assumption A.2, the privacy noise ξt is statistically independent of the gradient estimate¯gt and the fixed preconditionerP t

Quadratic Term (Penalty):We expand the squared norm of the update∆ t =−η t(Pt¯gt +ξ t): E[∥∆t∥2] =η 2 t E ∥Pt¯gt +ξ t∥2 (14) =η 2 t E[∥Pt¯gt∥2] +E[∥ξ t∥2] + 2E[⟨Pt¯gt, ξt⟩] (15) We

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
browse 1 citing papers

fields

cs.LG 1

years

2026 1

verdicts

UNVERDICTED 1

representative citing papers

DP-KFC: Data-Free Preconditioning for Privacy-Preserving Deep Learning

cs.LG · 2026-05-13 · unverdicted · novelty 6.0

DP-KFC approximates the Fisher Information Matrix for KFAC preconditioning via synthetic noise probes and modality frequency statistics, matching private-data performance without consuming privacy budget or introducing distribution shift.

citing papers explorer

Showing 1 of 1 citing paper.

  • DP-KFC: Data-Free Preconditioning for Privacy-Preserving Deep Learning cs.LG · 2026-05-13 · unverdicted · none · ref 6

    DP-KFC approximates the Fisher Information Matrix for KFAC preconditioning via synthetic noise probes and modality frequency statistics, matching private-data performance without consuming privacy budget or introducing distribution shift.