Interacting Particle Markov Chain Monte Carlo

By Tom Rainforth, C. A. Naesseth, Fredrik Lindsten, Brooks Paige, Jan-Willem van de Meent, Arnaud Doucet, Frank D. Wood. 2016

36 citations recorded by the source corpus.

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 noninteracting PMCMC samplers and a single PM-CMC 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.

Subfields: Artificial Intelligence, Statistics and Probability, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Frank D. Wood
2. Hongseok Yang
3. Jan-Willem van de Meent
4. David Tolpin
5. Yura N. Perov
6. Tuan Anh Le
7. Zoubin Ghahramani
8. Atilim Gunes Baydin

In Statistics and Probability

1. Hongseok Yang
2. Frank D. Wood
3. David Tolpin
4. Zoubin Ghahramani
5. Jan-Willem van de Meent
6. Tuan Anh Le
7. Atilim Gunes Baydin
8. Bradley Gram-Hansen

In Probabilistic & Bayesian Methods

1. Frank D. Wood
2. Hongseok Yang
3. Jan-Willem van de Meent
4. David Tolpin
5. Tuan Anh Le
6. Yura N. Perov
7. Zoubin Ghahramani
8. Atilim Gunes Baydin

Citations in the corpus

• Design and Implementation of Probabilistic Programming Language Anglican (2016, background): “New algorithms are being developed and added to Anglican (Tolpin et al. 2015a; van de Meent et al. 2016; Rainforth et al. 2016), as a part of ongoing research.”
• Inference Compilation and Universal Probabilistic Programming (2016, background): “…algorithms for universal probabilistic programming languages (Wingate et al., 2011; Ritchie et al., 2016a; Wood et al., 2014; Paige et al., 2014; Rainforth et al., 2016), we focus on algorithms in the importance sampling family in the context of which we will develop our scheme for amortized…”
• Turing: A Language for Flexible Probabilistic Inference (2018, background): “In fact, some MCMC operators are already contributed by community developers, such as PMMH [Andrieu et al., 2010] and IPMCMC [Rainforth et al., 2016] in Table 1.”
• Discontinuous Hamiltonian Monte Carlo for Probabilistic Programs (2018, methodology): “We compare against NUTS (Hoffman & Gelman, 2014) combined with Metropolis-within-Gibbs (NUTS+MwG) in PyMC3 (Salvatier et al., 2016) and Random-walk Lightweight Metropolis-Hastings (RMH) in Anglican (Wood et al., 2014).”