Skip to content
DraconiticResearch
Search Sources API
188,932 studies31 sourcesUpdated 37 min ago
PreprintCC BY 4.0arXiv

Sequince: a modular sequential Monte Carlo toolkit

  • Michael J. Williams
  • Christian E. A. Chapman-Bird
Open at sourcePDFDOICite

Abstract

Sequential Monte Carlo (SMC) methods approximate sequences of probability distributions using populations of weighted samples and are widely applicable, including in Bayesian inference and particle filtering. We present sequince, a modular Python toolkit for constructing SMC algorithms from reusable and extensible components. The toolkit provides modular steps for weighting, resampling, mutation, transformation, and control operations that are composed to implement different algorithms. It also includes prebuilt samplers for tempered SMC, persistent sampling and nested sampling via SMC. Its core components follow the Python Array API standard, supporting NumPy, JAX, PyTorch, and CuPy, enabling analyses on CPUs or GPUs. Random-number control and checkpointing support reproducible scientific workflows. This paper describes sequince v0.1.0 and its design as reusable infrastructure for methodological research and application of SMC methods.

Related

  1. Multigroup Implicit Monte Carlo Radiation Hydrodynamics on Moving Voronoi Meshes with STORM and RICH
    Maor Mizrachi, Barak Raveh, Elad Steinberg
    arXivPreprintCC BY 4.0Full textInstrumentation & methodsComputational physicsarXiv
  2. Markov Chain Monte Carlo for Bayesian Parametric Galaxy Modeling in LSST
    James J. Buchanan, Michael D. Schneider, Kerianne Pruett +1
    The Astrophysical JournalPeer-reviewedCC BY 4.0Full textGalaxiesCrossref
  3. Calculating the squeezed bispectrum in stochastic inflation by Monte Carlo simulation
    Koichi Miyamoto, Yuichiro Tada
    arXivPreprintarXiv licenceFull textCosmologyGravitation & general relativityarXiv
  4. A Delayed Rejection Reversible Jump Markov Chain Monte Carlo Method for Multi-Resolution Maps of the Stochastic Gravitational Wave Background with Pulsar Timing Array Data
    Taha T. Moursy, Nihan S. Pol
    arXivPreprintCC BY 4.0Full textHigh-energy astrophysicsInstrumentation & methodsarXiv
  5. Monte Carlo sampling of first-order QED processes in laser and pulsar plasmas
    Verneri Sarjomaa, Joonas Nättilä
    arXivPreprintCC BY 4.0Full textHigh-energy astrophysicsComputational physicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{williams2026sequince,
  title = {{Sequince: a modular sequential Monte Carlo toolkit}},
  author = {Michael J. Williams and Christian E. A. Chapman-Bird},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.39915},
  eprint = {2610.39915},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.39915},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - Sequince: a modular sequential Monte Carlo toolkit
AU  - Michael J. Williams
AU  - Christian E. A. Chapman-Bird
PY  - 2026
DA  - 2026/10/29
PB  - arXiv
DO  - 10.48550/arxiv.2610.39915
UR  - https://arxiv.org/abs/2610.39915
AB  - Sequential Monte Carlo (SMC) methods approximate sequences of probability distributions using populations of weighted samples and are widely applicable, including in Bayesian inference and particle filtering. We present sequince, a modular Python toolkit for constructing SMC algorithms from reusable and extensible components. The toolkit provides modular steps for weighting, resampling, mutation, transformation, and control operations that are composed to implement different algorithms. It also includes prebuilt samplers for tempered SMC, persistent sampling and nested sampling via SMC. Its core components follow the Python Array API standard, supporting NumPy, JAX, PyTorch, and CuPy, enabling analyses on CPUs or GPUs. Random-number control and checkpointing support reproducible scientific workflows. This paper describes sequince v0.1.0 and its design as reusable infrastructure for methodological research and application of SMC methods.
ER  -