Anchored Reversible Jump Sequential Monte Carlo

Abstract

Reversible Jump Markov Chain Monte Carlo (RJMCMC) is a principled framework for Bayesian model comparison, but its practical use is often limited due to the difficulty of designing between-model proposals that reach regions of high posterior probability. In addition, its inherently sequential nature limits efficient use of modern GPU architectures. We address both challenges by introducing Anchored Reversible Jump Sequential Monte Carlo. Our approach combines RJMCMC with Sequential Monte Carlo (SMC) to explore the model space in parallel. Importantly, the particle population enables effective between-model proposals without problem-specific knowledge: particles are used to construct kernel density approximations of the target distributions within each model, which are then used to generate trans-dimensional proposals. Numerical experiments show that the proposed method is computationally efficient and competitive with state-of-the-art RJMCMC approaches, demonstrating its potential for scalable Bayesian model comparison.

Fractional reversible chemical kinetics: Time-Driven Monte Carlo method and application to enzyme conformational dynamics
Nelson H. T. Lemes, Higor V. M. Ferreira
arXiv24 Oct 2026PreprintarXiv licenceFull textMathematical physicsarXiv
Differentiable Systematic Resampling for Variational Sequential Monte Carlo
Fredrik Cumlin, Saikat Chatterjee
arXiv28 Oct 2026PreprintCC BY 4.0Full textStatisticsarXiv
Sequince: a modular sequential Monte Carlo toolkit
Michael J. Williams, Christian E. A. Chapman-Bird
arXiv29 Oct 2026PreprintCC BY 4.0Full textInstrumentation & methods StatisticsarXiv
Parameter estimation in Conditional Sequential Monte Carlo algorithms through Particle Learning
Alfonso Diz-Lois Palomares, Geir Storvik
arXiv28 Aug 2026PreprintCC BY-NC-ND 4.0Full textStatisticsarXiv
Steering Diffusion Models to Rare Events with Sequential Monte Carlo
Aavash Subedi, Tim Reichelt, Christopher Williams +3
arXiv13 Oct 2026PreprintCC BY 4.0Full textStatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{assen2026anchored,
  title = {{Anchored Reversible Jump Sequential Monte Carlo}},
  author = {Janna van Assen and Max Hinne},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.09930},
  eprint = {2610.09930},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.09930},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Anchored Reversible Jump Sequential Monte Carlo
AU  - Janna van Assen
AU  - Max Hinne
PY  - 2026
DA  - 2026/10/07
PB  - arXiv
DO  - 10.48550/arxiv.2610.09930
UR  - https://arxiv.org/abs/2610.09930
AB  - Reversible Jump Markov Chain Monte Carlo (RJMCMC) is a principled framework for Bayesian model comparison, but its practical use is often limited due to the difficulty of designing between-model proposals that reach regions of high posterior probability. In addition, its inherently sequential nature limits efficient use of modern GPU architectures. We address both challenges by introducing Anchored Reversible Jump Sequential Monte Carlo. Our approach combines RJMCMC with Sequential Monte Carlo (SMC) to explore the model space in parallel. Importantly, the particle population enables effective between-model proposals without problem-specific knowledge: particles are used to construct kernel density approximations of the target distributions within each model, which are then used to generate trans-dimensional proposals. Numerical experiments show that the proposed method is computationally efficient and competitive with state-of-the-art RJMCMC approaches, demonstrating its potential for scalable Bayesian model comparison.
ER  -

Abstract

Related

Cite