Exponential-Wrapped Mechanisms: Differential Privacy on Hadamard Manifolds Made Practical

ICLR 2026 Conference Submission22531 Authors

Keywords: Differential Privacy, Riemannian Manifold, Hadamard manifolds, Fréchet mean, SPDM space

TL;DR: e propose fast, general mechanisms for differential privacy on Hadamard manifolds using exponential wrapping, achieving DP, ADP, GDP, and RDP without MCMC.

Abstract: We propose a general and computationally efficient framework for achieving differential privacy (DP) on Hadamard manifolds, which are complete and simply connected Riemannian manifolds with non-positive curvature. Leveraging the Cartan-Hadamard theorem, we introduce Exponential-Wrapped Laplace and Gaussian mechanisms that achieve $\epsilon$-DP, $(\epsilon, \delta)$-DP, Gaussian DP (GDP), and Rényi DP (RDP) without relying on computationally intensive MCMC sampling. Our methods operate entirely within the intrinsic geometry of the manifold, ensuring both theoretical soundness and practical scalability. We derive utility bounds for privatized Fréchet means and demonstrate superior utility and runtime performances on both synthetic data and real-world data in the space of symmetric positive definite matrices (SPDM) equipped with three different metrics. To our knowledge, this work constitutes the first unified extension of multiple DP notions to general Hadamard manifolds with practical and scalable implementations.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Submission Number: 22531