Geodesic Slice Sampler for Multimodal Distributions with Strong Curvature

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Williams, Bernardo, Yu, Hanlin, Luu, Hoang Phuc Hau, Arvanitidis, Georgios, Klami, Arto
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916794711670784
author Williams, Bernardo
Yu, Hanlin
Luu, Hoang Phuc Hau
Arvanitidis, Georgios
Klami, Arto
author_facet Williams, Bernardo
Yu, Hanlin
Luu, Hoang Phuc Hau
Arvanitidis, Georgios
Klami, Arto
contents Traditional Markov Chain Monte Carlo sampling methods often struggle with sharp curvatures, intricate geometries, and multimodal distributions. Slice sampling can resolve local exploration inefficiency issues, and Riemannian geometries help with sharp curvatures. Recent extensions enable slice sampling on Riemannian manifolds, but they are restricted to cases where geodesics are available in a closed form. We propose a method that generalizes Hit-and-Run slice sampling to more general geometries tailored to the target distribution, by approximating geodesics as solutions to differential equations. Our approach enables the exploration of the regions with strong curvature and rapid transitions between modes in multimodal distributions. We demonstrate the advantages of the approach over challenging sampling problems.
format Preprint
id arxiv_https___arxiv_org_abs_2502_21190
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Geodesic Slice Sampler for Multimodal Distributions with Strong Curvature
Williams, Bernardo
Yu, Hanlin
Luu, Hoang Phuc Hau
Arvanitidis, Georgios
Klami, Arto
Machine Learning
Traditional Markov Chain Monte Carlo sampling methods often struggle with sharp curvatures, intricate geometries, and multimodal distributions. Slice sampling can resolve local exploration inefficiency issues, and Riemannian geometries help with sharp curvatures. Recent extensions enable slice sampling on Riemannian manifolds, but they are restricted to cases where geodesics are available in a closed form. We propose a method that generalizes Hit-and-Run slice sampling to more general geometries tailored to the target distribution, by approximating geodesics as solutions to differential equations. Our approach enables the exploration of the regions with strong curvature and rapid transitions between modes in multimodal distributions. We demonstrate the advantages of the approach over challenging sampling problems.
title Geodesic Slice Sampler for Multimodal Distributions with Strong Curvature
topic Machine Learning
url https://arxiv.org/abs/2502.21190