Jul 27, 2021 · We introduce a gradient-descent-based tool to learn truncated noise for additive mechanisms with strong utility bounds while simultaneously optimizing for ...

Differentially private (DP) mechanisms face the challenge of providing accurate results while protecting their inputs: the privacy-utility trade-off.

Learning Numeric Optimal Differentially Private Truncated Additive Mechanisms. David M. Sommer, Lukas Abfalterer, Sheila Zingg, Esfandiar Mohammadi.

David M. Sommer, Lukas Abfalterer, Sheila Zingg, Esfandiar Mohammadi: Learning Numeric Optimal Differentially Private Truncated Additive Mechanisms.

Learning Numeric Optimal Differentially Private Truncated Additive Mechanisms ... An additive mechanism with truncated noise (i. e., with bounded range) can offer ...

Learning numeric optimal differentially private truncated additive mechanisms. DM Sommer, L Abfalterer, S Zingg, E Mohammadi. arXiv preprint arXiv:2107.12957, ...

Differential privacy offers clear and strong quantitative guarantees for privacy mechanisms, but it assumes an attacker that knows all but one records of the ...

This work considers precise composition bounds of the overall privacy loss for exponential mechanisms, one of the fundamental classes of mechanisms in DP ...

Learning Numeric Optimal Differentially Private Truncated Additive Mechanisms · Schrödinger Mechanisms: Optimal Differential Privacy Mechanisms for Small ...

Aug 30, 2024 · We introduce a differentially private set-based estimator leveraging a truncated additive mechanism with a numerically optimized noise ...