Interacting Particle Markov Chain Monte Carlo
This addresses a bottleneck in Bayesian inference for researchers and practitioners by enhancing computational efficiency in sampling methods, though it is incremental as it builds on existing PMCMC frameworks.
The paper tackles the problem of improving mixing rates in Markov chain Monte Carlo sampling by introducing interacting particle Markov chain Monte Carlo (iPMCMC), which uses an interacting pool of sequential Monte Carlo samplers, and shows significant improvements in mixing rates compared to non-interacting methods and single samplers with equivalent resources.
We introduce interacting particle Markov chain Monte Carlo (iPMCMC), a PMCMC method based on an interacting pool of standard and conditional sequential Monte Carlo samplers. Like related methods, iPMCMC is a Markov chain Monte Carlo sampler on an extended space. We present empirical results that show significant improvements in mixing rates relative to both non-interacting PMCMC samplers, and a single PMCMC sampler with an equivalent memory and computational budget. An additional advantage of the iPMCMC method is that it is suitable for distributed and multi-core architectures.