A General Metric for Riemannian Manifold Hamiltonian Monte Carlo

Abstract

Markov Chain Monte Carlo (MCMC) is an invaluable means of inference with complicated models, and Hamiltonian Monte Carlo, in particular Riemannian Manifold Hamiltonian Monte Carlo (RMHMC), has demonstrated impressive success in many challenging problems. Current RMHMC implementations, however, rely on a Riemannian metric that limits their application to analytically-convenient models. In this paper I propose a new metric for RMHMC without these limitations and verify its success on a distribution that emulates many hierarchical and latent models.

13 pages, 10 images

Cited by in corpus (11)

• A Review of automatic differentiation and its efficient implementation
• A Survey of Stochastic Simulation and Optimization Methods in Signal Processing
• On the Geometric Ergodicity of Hamiltonian Monte Carlo
• Langevin and Hamiltonian based Sequential MCMC for Efficient Bayesian Filtering in High-dimensional Spaces
• Geometric variational inference
• Modified Hamiltonian Monte Carlo for Bayesian inference
• Trajectory Based RFI Subtraction and Calibration for Radio Interferometry
• Differentiable programming for particle physics simulations
• Truncated Log-concave Sampling with Reflective Hamiltonian Monte Carlo
• High-dimensional Nonlinear Bayesian Inference of Poroelastic Fields from Pressure Data
• Delayed rejection Hamiltonian Monte Carlo for sampling multiscale distributions