Tight Group-Level DP Guarantees for DP-SGD with Sampling via Mixture of Gaussians Mechanisms
Not on the map. Why?
Abstract (arXiv)
We give a procedure for computing group-level (ε, δ)-DP guarantees for DP-SGD, when using Poisson sampling or fixed batch size sampling. Up to discretization errors in the implementation, the DP guarantees computed by this procedure are tight (assuming we release every intermediate iterate).
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 8 of its 10 references that are in the map. First posted on arXiv 17 Jan 2024.
Builds on
8 of 10 references.
References · 8
- 2021Numerical Composition of Differential Privacycited by 174
- 2018Privacy Amplification by Iterationcited by 97
- 2021Optimal Accounting of Differential Privacy via Characteristic Functioncited by 83
- 2021Differential Privacy Dynamics of Langevin Diffusion and Noisy Gradient Descentcited by 71
- 2022Privacy of Noisy Stochastic Gradient Descent: More Iterations without More Privacy Losscited by 60
- 2022Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributionscited by 50
- 2023(Amplified) Banded Matrix Factorization: A unified approach to private trainingcited by 49
- 2020Tight Differential Privacy for Discrete-Valued Mechanisms and for the Subsampled Gaussian Mechanism Using FFTcited by 43
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.