Arbitrary Marginal Neural Ratio Estimation for Simulation-based\n Inference

2021/10/01 by François Rozet, Gilles Louppe, Rozet, François +1 · 1 citation
Earth and Planetary Sciences · Engineering · Mathematics · Physics and Astronomy · #FOS: Computer and information sciences #FOS: Physical sciences #Instrumentation and Methods for Astrophysics (astro-ph.IM) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Model Reduction and Neural Networks #Pulsars and Gravitational Waves Research #Reservoir Engineering and Simulation Methods #Seismic Imaging and Inversion Techniques #Statistical and numerical algorithms

paper · pdf · doi:10.48550/arxiv.2110.00449

openalex publication_date 2021/10/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In many areas of science, complex phenomena are modeled by stochastic\nparametric simulators, often featuring high-dimensional parameter spaces and\nintractable likelihoods. In this context, performing Bayesian inference can be\nchallenging. In this work, we present a novel method that enables amortized\ninference over arbitrary subsets of the parameters, without resorting to\nnumerical integration, which makes interpretation of the posterior more\nconvenient. Our method is efficient and can be implemented with arbitrary\nneural network architectures. We demonstrate the applicability of the method on\nparameter inference of binary black hole systems from gravitational waves\nobservations.\n

Citations

• Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Monte Carlo sampling methods using Markov chains and their applications
• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• Observation of Gravitational Waves from a Binary Black Hole Merger
• SGDR: Stochastic Gradient Descent with Warm Restarts
• Dynamic nested sampling: an improved algorithm for parameter estimation and evidence calculation
• Learning about an exponential amount of conditional distributions
• dynesty: a dynamic nested sampling package for estimating Bayesian posteriors and evidences
• Automatic Posterior Transformation for Likelihood-Free Inference
• Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows

Cited by

• Fair Universe Higgs Uncertainty Challenge

Related

• Solving high-dimensional parameter inference: marginal posterior densities & Moment Networks
• Tuning neural posterior estimation for gravitational wave inference
• Simulation-based inference for stochastic gravitational wave background data analysis
• Lightning-Fast Gravitational Wave Parameter Inference through Neural Amortization
• Simulation-Based Inference: A Practical Guide