Generalizing the No-U-Turn Sampler to Riemannian Manifolds

Abstract

Hamiltonian Monte Carlo provides efficient Markov transitions at the expense of introducing two free parameters: a step size and total integration time. Because the step size controls discretization error it can be readily tuned to achieve certain accuracy criteria, but the total integration time is left unconstrained. Recently Hoffman and Gelman proposed a criterion for tuning the integration time in certain systems with their No U-Turn Sampler, or NUTS. In this paper I investigate the dynamical basis for the success of NUTS and generalize it to Riemannian Manifold Hamiltonian Monte Carlo.

8 pages, 6 figures

Cited by in corpus (12)

• Optimizing The Integrator Step Size for Hamiltonian Monte Carlo
• Spatial Bayesian hierarchical modeling of precipitation extremes over a large domain
• Modified Hamiltonian Monte Carlo for Bayesian inference
• Identifying the Optimal Integration Time in Hamiltonian Monte Carlo
• Scaling Hamiltonian Monte Carlo Inference for Bayesian Neural Networks with Symmetric Splitting
• Introducing an Explicit Symplectic Integration Scheme for Riemannian Manifold Hamiltonian Monte Carlo
• Bayesian Inference for Randomized Benchmarking Protocols
• Hamiltonian Monte Carlo on Symmetric and Homogeneous Spaces via Symplectic Reduction
• Antithetic Riemannian Manifold And Quantum-Inspired Hamiltonian Monte Carlo
• Hawai'i Supernova Flows: A Peculiar Velocity Survey Using Over a Thousand Supernovae in the Near-Infrared
• Geodesic Lagrangian Monte Carlo over the space of positive definite matrices: with application to Bayesian spectral density estimation
• A Mathematical Walkthrough and Discussion of the Free Energy Principle