Diagnosing Suboptimal Cotangent Disintegrations in Hamiltonian Monte Carlo
arXiv:1604.00695
Abstract
When properly tuned, Hamiltonian Monte Carlo scales to some of the most challenging high-dimensional problems at the frontiers of applied statistics, but when that tuning is suboptimal the performance leaves much to be desired. In this paper I show how suboptimal choices of one critical degree of freedom, the cotangent disintegration, manifest in readily observed diagnostics that facilitate the robust application of the algorithm.
17 pages, 9 figures
References in corpus (5)
- A General Framework for the Parametrization of Hierarchical Models
- On the Geometric Ergodicity of Hamiltonian Monte Carlo
- Optimizing The Integrator Step Size for Hamiltonian Monte Carlo
- The Geometric Foundations of Hamiltonian Monte Carlo
- Identifying the Optimal Integration Time in Hamiltonian Monte Carlo
Cited by in corpus (11)
- A Conceptual Introduction to Hamiltonian Monte Carlo
- Bayesian aggregation of average data: An application in drug development
- Almanac: Weak Lensing power spectra and map inference on the masked sphere
- Hierarchical Bayesian CMB Component Separation with the No-U-Turn Sampler
- Precision Ephemerides for Gravitational-wave Searches -- IV: Corrected and refined ephemeris for Scorpius X-1
- Sensitivity Analyses of Exoplanet Occurrence Rates from Kepler and Gaia
- Almanac: MCMC-based signal extraction of power spectra and maps on the sphere
- Selection of inverse gamma and half-t priors for hierarchical models: sensitivity and recommendations
- Forecasting Fertility with Parametric Mixture Models
- Spatial modelling of COVID-19 incident cases using Richards' curve: an application to the Italian regions
- Skills to not fall behind in school