PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 13, 2026Journal of the Royal Statistical Society Series B (Statistical Methodology)0 citations

Sampling from Multimodal Distributions Using Tempered Hamiltonian Monte Carlo

Sampling from high-dimensional, multimodal distributions using automatically tuned, tempered Hamiltonian Monte Carlo

View Full Paper

Authors

JPJoon Ha Park

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates improved sampling from high-dimensional multimodal distributions, indicating enhanced efficiency for Bayesian inference.

Key Points

  • This study aims to improve sampling efficiency from high-dimensional, strongly multimodal distributions using a new approach that combines tempering with Hamiltonian Monte Carlo.
  • Developed a tempered Hamiltonian Monte Carlo (THMC) algorithm with automatic tuning.
  • Simulated dynamics of a time-varying Hamiltonian with temperature changes to explore distributions.
  • Compared THMC's performance against adaptive parallel tempering and tempered sequential Monte Carlo.
  • THMC scales more effectively with dimension than adaptive parallel tempering (p<0.01).
  • Showed improved sampling efficiency from strongly multimodal posterior distributions.
  • Demonstrated effective exploration of low-density regions and guidance towards local modes.
Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Joon Ha Park (2026) studied this question.

synapsesocial.com/papers/6a2cf500faef96ed7f057216https://doi.org/10.1093/jrsssb/qkag080
Ask AI
Helpful
Bookmark
Share
View Full Paper