Initializing adaptive importance sampling with Markov chains

Abstract

Adaptive importance sampling is a powerful tool to sample from complicated target densities, but its success depends sensitively on the initial proposal density. An algorithm is presented to automatically perform the initialization using Markov chains and hierarchical clustering. The performance is checked on challenging multimodal examples in up to 20 dimensions and compared to results from nested sampling. Our approach yields a proposal that leads to rapid convergence and accurate estimation of overall normalization and marginal distributions.

17 pages, 4 figures

References in corpus (1)

• Estimation of cosmological parameters using adaptive importance sampling

Cited by in corpus (11)

• Importance Nested Sampling and the MultiNest Algorithm
• Comprehensive Bayesian Analysis of Rare (Semi)leptonic and Radiative B Decays
• Nested sampling for physical scientists
• Layered Adaptive Importance Sampling
• EOS -- A Software for Flavor Physics Phenomenology
• (MC)**3 -- a Multi-Channel Markov Chain Monte Carlo algorithm for phase-space sampling
• Nested sampling cross-checks using order statistics
• MCMC-driven importance samplers
• Black-box Importance Sampling
• Estimation and prediction for spatial generalized linear mixed models with parametric links via reparameterized importance sampling
• EOS -- A Software for Flavor Physics Phenomenology