ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1805.05289
13
2
v1v2 (latest)

Note on the geodesic Monte Carlo

14 May 2018
Andrew J Holbrook
ArXiv (abs)PDFHTML
Abstract

Geodesic Monte Carlo (gMC) is a powerful algorithm for Bayesian inference on non-Euclidean manifolds. The original gMC algorithm was cleverly derived in terms of its progenitor, the Riemannian manifold Hamiltonian Monte Carlo (RMHMC). Here, it is shown that alternative and theoretically simpler derivations are available in which the original algorithm is a special case of two general classes of algorithms characterized by non-trivial mass matrices. The proposed derivations work entirely in embedding coordinates and thus clarify the original algorithm as applied to manifolds embedded in Euclidean space.

View on arXiv
Comments on this paper