Massively Parallel Dimension Independent Adaptive Metropolis

Massively Parallel Dimension Independent Adaptive Metropolis

This work considers black-box Bayesian inference over high-dimensional parameter spaces. The well-known and widely respected adaptive Metropolis (AM) algorithm is extended herein to asymptotically scale uniformly with respect to the underlying parameter dimension, by respecting the variance, for Gau...

Full description

Bibliographic Details
Main Author: Chen, Yuxin
Other Authors: Keyes, David E.
Language:en
Published: 2015
Subjects:
Markov chain Monte Carlo
bayesian interference
adaptive metropolis
metropolis hastings
parallele processing
GPM acceleration
Online Access:http://hdl.handle.net/10754/552902

Internet

http://hdl.handle.net/10754/552902

Similar Items