Husmeier, D. (2011) Contribution to the discussion on Riemann manifold Hamiltonian Monte Carlo methods. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 73(2), pp. 184-185. (doi: 10.1111/j.1467-9868.2010.00765.x)
Full text not currently available from Enlighten.
Abstract
By combining concepts from physics (Hamiltonian dynamics) with Riemann geometry (metric tensor), the authors have made a highly important contribution to current research on MCMC, which in my opinion will be as seminal as Green (1995). In this brief note, I shall discuss how mixing in the presence of multimodality may be further improved. For illustration, consider the Gaussian mixture
Item Type: | Articles |
---|---|
Status: | Published |
Refereed: | Yes |
Glasgow Author(s) Enlighten ID: | Husmeier, Professor Dirk |
Authors: | Husmeier, D. |
College/School: | College of Science and Engineering > School of Mathematics and Statistics > Statistics |
Journal Name: | Journal of the Royal Statistical Society: Series B (Statistical Methodology) |
Journal Abbr.: | J. R. Stat. Soc. Series B. Stat. Methodol. |
ISSN: | 1369-7412 |
ISSN (Online): | 1467-9868 |
University Staff: Request a correction | Enlighten Editors: Update this record