Slice sampling covariance hyperparameters of latent Gaussian models
arXiv:1006.0868
Abstract
The Gaussian process (GP) is a popular way to specify dependencies between random variables in a probabilistic model. In the Bayesian framework the covariance structure can be specified using unknown hyperparameters. Integrating over these hyperparameters considers different possible explanations for the data when making predictions. This integration is often performed using Markov chain Monte Carlo (MCMC) sampling. However, with non-Gaussian observations standard hyperparameter sampling approaches require careful tuning and may converge slowly. In this paper we present a slice sampling approach that requires little tuning while mixing well in both strong- and weak-data regimes.
9 pages, 4 figures, 4 algorithms. Minor corrections to previous version. This version to appear in Advances in Neural Information Processing Systems (NIPS) 23, 2010
References in corpus (1)
Cited by in corpus (9)
- Practical Bayesian Optimization of Machine Learning Algorithms
- MCMC for Variationally Sparse Gaussian Processes
- Raiders of the Lost Architecture: Kernels for Bayesian Optimization in Conditional Parameter Spaces
- Probabilistic prediction of neurological disorders with a statistical assessment of neuroimaging data modalities
- Neutron Transmission Strain Tomography for Non-Constant Stress-Free Lattice Spacing
- Bayesian Optimization for Policy Search via Online-Offline Experimentation
- Space and circular time log Gaussian Cox processes with application to crime event data
- A Probabilistic Approach to Nonparametric Local Volatility
- Learning Insulin-Glucose Dynamics in the Wild