English

Algorithmic Aspects of the Log-Laplace Transform and a Non-Euclidean Proximal Sampler

Data Structures and Algorithms 2025-12-24 v3 Cryptography and Security Machine Learning Probability Computation

Abstract

The development of efficient sampling algorithms catering to non-Euclidean geometries has been a challenging endeavor, as discretization techniques which succeed in the Euclidean setting do not readily carry over to more general settings. We develop a non-Euclidean analog of the recent proximal sampler of [LST21], which naturally induces regularization by an object known as the log-Laplace transform (LLT) of a density. We prove new mathematical properties (with an algorithmic flavor) of the LLT, such as strong convexity-smoothness duality and an isoperimetric inequality, which are used to prove a mixing time on our proximal sampler matching [LST21] under a warm start. As our main application, we show our warm-started sampler improves the value oracle complexity of differentially private convex optimization in p\ell_p and Schatten-pp norms for p[1,2]p \in [1, 2] to match the Euclidean setting [GLL22], while retaining state-of-the-art excess risk bounds [GLLST23]. We find our investigation of the LLT to be a promising proof-of-concept of its utility as a tool for designing samplers, and outline directions for future exploration.

Keywords

Cite

@article{arxiv.2302.06085,
  title  = {Algorithmic Aspects of the Log-Laplace Transform and a Non-Euclidean Proximal Sampler},
  author = {Sivakanth Gopi and Yin Tat Lee and Daogao Liu and Ruoqi Shen and Kevin Tian},
  journal= {arXiv preprint arXiv:2302.06085},
  year   = {2025}
}

Comments

Fixed error in previous version, main result weakened by a quadratic factor (see discussion in Section 1.4)