Markov Chain Monte Carlo posterior sampling with the Hamiltonian method

Proceedings of SPIE - The International Society for Optical Engineering

Volumn 4322, Issue 1, 2001, Pages 456-467

  Hanson, K M ^a  

^a LOS ALAMOS NATIONAL LABORATORY   (United States)

Author keywords

Bayesian analysis; Convergence test; Hamiltonian method; Hybrid MCMC; Markov Chain Monte Carlo; Metropolis method; Posterior distribution; Statistical efficiency; Uncertainty estimation

Indexed keywords

ALGORITHMS; COMPUTER SIMULATION; CONVERGENCE OF NUMERICAL METHODS; HAMILTONIANS; KINETIC ENERGY; MARKOV PROCESSES; MONTE CARLO METHODS; POTENTIAL ENERGY; PROBABILITY DENSITY FUNCTION; PROBABILITY DISTRIBUTIONS;

BAYESIAN ANALYSIS;

IMAGE RECONSTRUCTION;

EID: 0034847223     PISSN: 0277786X     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1117/12.431119     Document Type: Conference Paper

References (20)

1. Bayesian computation and stochastic systems
2. Efficient metropolis jumping rules
3. Posterior sampling with improved efficiency
4. Hybrid Monte Carlo
5. Molecular dynamics simulations at constant pressure and/or temperature
6. A generalized guided Monte Carlo algorithm
7. Inversion based on computational simulations
8. Construction of the adjoint of the HOPE OGCM
9. Three-dimensional reconstructions from low-count SPECT data using deformable models
10. Equations of state calculations by fast computing machine
11. Implementing MCMC
12. Inference from iterative simulation using multiple sequences (with discussion)
13. Adjoint differentiation of hydrodynamic codes