English

Truncated Marginal Neural Ratio Estimation

Machine Learning 2021-10-27 v2 Instrumentation and Methods for Astrophysics Machine Learning High Energy Physics - Phenomenology

Abstract

Parametric stochastic simulators are ubiquitous in science, often featuring high-dimensional input parameters and/or an intractable likelihood. Performing Bayesian parameter inference in this context can be challenging. We present a neural simulation-based inference algorithm which simultaneously offers simulation efficiency and fast empirical posterior testability, which is unique among modern algorithms. Our approach is simulation efficient by simultaneously estimating low-dimensional marginal posteriors instead of the joint posterior and by proposing simulations targeted to an observation of interest via a prior suitably truncated by an indicator function. Furthermore, by estimating a locally amortized posterior our algorithm enables efficient empirical tests of the robustness of the inference results. Since scientists cannot access the ground truth, these tests are necessary for trusting inference in real-world applications. We perform experiments on a marginalized version of the simulation-based inference benchmark and two complex and narrow posteriors, highlighting the simulator efficiency of our algorithm as well as the quality of the estimated marginal posteriors.

Keywords

Cite

@article{arxiv.2107.01214,
  title  = {Truncated Marginal Neural Ratio Estimation},
  author = {Benjamin Kurt Miller and Alex Cole and Patrick Forré and Gilles Louppe and Christoph Weniger},
  journal= {arXiv preprint arXiv:2107.01214},
  year   = {2021}
}

Comments

10 pages. 27 pages with references and supplemental material. Implementation of experiments at https://github.com/bkmi/tmnre/. Ready-to-use implementation of underlying algorithm at https://github.com/undark-lab/swyft/. Accepted at NeurIPS 2021

R2 v1 2026-06-24T03:51:11.534Z