Divide-and-Conquer Neural Bayes Estimation for Spatial and Spatiotemporal Point Process Models

Abstract

Spatial and spatiotemporal point process models are widely used to analyze event data across many scientific disciplines, but likelihood-based Bayesian inference is often computationally prohibitive due to intractable likelihoods, high-dimensional latent structures, and dependence on event history. Existing approximate and likelihood-free methods mitigate some of these challenges, but often at the cost of reduced statistical efficiency, imperfect uncertainty quantification, or information loss from reliance on predefined summary statistics. A promising recent approach involves neural networks, but it has only been explored in purely spatial or temporal settings and requires a computationally expensive training stage for large domains or dense point patterns. Here, we propose a divide-and-conquer neural Bayes estimation framework for spatial and spatiotemporal point process models. Our method operates directly on raw point pattern realizations using transformer-based architectures on partitions of the spatiotemporal domain, together with an attention mechanism to optimally combine locally extracted summaries. Simulation studies demonstrate reliable inference with improved scalability relative to existing general-purpose neural inference approaches and classical likelihood-based methods. We further demonstrate the practical applicability of our approach through an analysis of whale call data in Cape Cod Bay, MA, where, once trained, the estimator yields parameter estimates and credible intervals at negligible computational cost.

Scalable Bayesian Inference for Time Series via Divide and Conquer
Rihui Ou, Lachlan Astfalck, Deborshee Sen +1
Journal of the American Statistical Association21 Jul 2026Peer-reviewedCC BY6 citedStatisticsOpenAlex
Two moments suffice for Bessel approximation: conditioned differences, contingency tables, and probabilistic divide-and-conquer
Stephen DeSalvo
arXiv1 Nov 2026PreprintarXiv licenceFull textCombinatorics & discrete mathematics ProbabilityarXiv
Scaling Spatially Varying Distributed Lag Nonlinear Models for Nationwide Small-Area Data: A Divide-and-Conquer Approach
Sara Rutten, Thomas Neyens, Elisa Duarte +1
arXiv16 Oct 2026PreprintCC BY-NC-ND 4.0Full textStatisticsarXiv
Spectral divide-and-conquer MCMC for long stationary time series
Zixuan Wang, Matias Quiroz, Feng Li +2
arXiv21 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Gaussian Process Decorrelation for Spatiotemporal Deep Learning-Based Snow Water Equivalent Prediction
Colin Fenster, Adrienne Marshall, Soutir Bandyopadhyay +1
arXiv1 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{kang2026divideandconquer,
  title = {{Divide-and-Conquer Neural Bayes Estimation for Spatial and Spatiotemporal Point Process Models}},
  author = {Bokgyeong Kang and Andrew Zammit-Mangion and Erin M. Schliep and Robert S. Schick},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.27619},
  eprint = {2610.27619},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.27619},
}

RIS

TY  - UNPB
TI  - Divide-and-Conquer Neural Bayes Estimation for Spatial and Spatiotemporal Point Process Models
AU  - Bokgyeong Kang
AU  - Andrew Zammit-Mangion
AU  - Erin M. Schliep
AU  - Robert S. Schick
PY  - 2026
DA  - 2026/10/20
PB  - arXiv
DO  - 10.48550/arxiv.2610.27619
UR  - https://arxiv.org/abs/2610.27619
AB  - Spatial and spatiotemporal point process models are widely used to analyze event data across many scientific disciplines, but likelihood-based Bayesian inference is often computationally prohibitive due to intractable likelihoods, high-dimensional latent structures, and dependence on event history. Existing approximate and likelihood-free methods mitigate some of these challenges, but often at the cost of reduced statistical efficiency, imperfect uncertainty quantification, or information loss from reliance on predefined summary statistics. A promising recent approach involves neural networks, but it has only been explored in purely spatial or temporal settings and requires a computationally expensive training stage for large domains or dense point patterns. Here, we propose a divide-and-conquer neural Bayes estimation framework for spatial and spatiotemporal point process models. Our method operates directly on raw point pattern realizations using transformer-based architectures on partitions of the spatiotemporal domain, together with an attention mechanism to optimally combine locally extracted summaries. Simulation studies demonstrate reliable inference with improved scalability relative to existing general-purpose neural inference approaches and classical likelihood-based methods. We further demonstrate the practical applicability of our approach through an analysis of whale call data in Cape Cod Bay, MA, where, once trained, the estimator yields parameter estimates and credible intervals at negligible computational cost.
ER  -

Abstract

Related

Cite