Projects per year
Abstract
Sampling from hierarchical Bayesian models is often difficult for MCMC methods, because of the strong correlations between the model parameters and the hyperparameters. Recent Riemannian manifold Hamiltonian Monte Carlo (RMHMC) methods have significant potential advantages in this setting, but are computationally expensive. We introduce a new RMHMC method, which we call semiseparable Hamiltonian Monte Carlo, which uses a specially designed mass matrix that allows the joint Hamiltonian over model parameters and hyperparameters to decompose into two simpler Hamiltonians. This structure is exploited by a new integrator which we call the alternating blockwise leapfrog algorithm. The resulting method can mix faster than simpler Gibbs sampling while being simpler and more efficient than previous instances of RMHMC.
Original language  English 

Title of host publication  Advances in Neural Information Processing Systems 27 
Editors  Z. Ghahramani, M. Welling, C. Cortes, N.D. Lawrence, K.Q. Weinberger 
Publisher  Curran Associates Inc 
Pages  1018 
Number of pages  9 
Publication status  Published  2014 
Fingerprint
Dive into the research topics of 'SemiSeparable Hamiltonian Monte Carlo for Inference in Bayesian Hierarchical Models'. Together they form a unique fingerprint.Projects
 1 Finished

Fast, locally adaptive interference for machine learning in graphical models
1/10/11 → 30/09/14
Project: Research