The Multiple-Try Method and Local Optimization in Metropolis Sampling

@inproceedings{Liu2000TheMM,
  title={The Multiple-Try Method and Local Optimization in Metropolis Sampling},
  author={Jun S. Liu and Faming Liang and Wing Hung Wong},
  year={2000}
}
  • Jun S. Liu, Faming Liang, Wing Hung Wong
  • Published 2000
  • Mathematics
  • Abstract This article describes a new Metropolis-like transition rule, the multiple-try Metropolis, for Markov chain Monte Carlo (MCMC) simulations. By using this transition rule together with adaptive direction sampling, we propose a novel method for incorporating local optimization steps into a MCMC sampler in continuous state-space. Numerical studies show that the new method performs significantly better than the traditional Metropolis-Hastings (M-H) sampler. With minor tailoring in using… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 247 CITATIONS

    Mitsuba 2: a retargetable forward and inverse renderer

    VIEW 5 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Paired-move multiple-try stochastic search for Bayesian variable selection

    VIEW 4 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Algorithms and architectures for MCMC acceleration in FPGAs

    VIEW 4 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Issues in the Multiple Try Metropolis mixing

    VIEW 15 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Topics in Modern Bayesian Computation

    VIEW 19 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Additive Gaussian Process Regression

    VIEW 8 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Interacting multiple try algorithms with different proposal distributions

    VIEW 13 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Bayesian Inference on Structure Identification

    VIEW 10 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    FILTER CITATIONS BY YEAR

    2000
    2020

    CITATION STATISTICS

    • 39 Highly Influenced Citations

    • Averaged 21 Citations per year from 2017 through 2019