Truncated Marginal Neural Ratio Estimation

TL;DR

This paper introduces a neural simulation-based inference method that efficiently estimates marginal posteriors, allows targeted simulations, and provides empirical robustness tests, addressing challenges in high-dimensional and intractable likelihood scenarios.

Contribution

The proposed algorithm uniquely combines simulation efficiency with empirical posterior testing by estimating marginal posteriors and using targeted, truncated simulations.

Findings

Demonstrates efficiency on benchmark and complex posteriors

Provides reliable empirical tests of inference robustness

Achieves accurate marginal posterior estimation

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.