Amortized ratio-estimation importance sampling and localized simulation-based calibration for intractable likelihoods
Abstract
We consider simulation-based Bayesian inference (SBI) for the parameters of models with intractable likelihoods but tractable forward simulation. Building on Gaussian mixtures-of-experts surrogates, as a first contribution we develop a sequential procedure to estimate the posterior, incorporating an importance sampling step that introduces a locally amortized likelihood-to-evidence ratio estimator. The local amortization allows the ratio estimator to be evaluated at new parameter--data pairs from a local (data-dependent) design, without re-estimation, which is particularly convenient for our second contribution. The second contribution is localized simulation-based calibration (SBC). Localized SBC provides a graphical diagnostic of calibration over neighborhoods that are deliberately broader than the available posterior, at low additional computational cost. Rather than repeatedly running the full inference procedure for each simulated pseudo-observation, localized SBC fits the local surrogate and ratio estimator only once for each local design, and reuses them across all pseudo-observations. This approach is particularly appealing for SBI applications. We evaluate both contributions on three case studies, including a real-data epidemiological application.
Related
- A Proxy-likelihood Estimator for Multivariate Extremes Models with Intractable Likelihoods
- Efficient Bootstrap Calibration of Generalized Posteriors with Importance Reweighting
- LUCID: Local Understanding of Covariate Importance with Conditional Density Estimation
- Learning under Target Shift: Optimal Density Ratio Estimation and Importance-Weighted Regression
- Pairwise Distortion Distribution with Graph-Based Importance Sampling for Quantization Error and Intrinsic Dimension Estimation
BibTeX
@misc{picchini2026amortized,
title = {{Amortized ratio-estimation importance sampling and localized simulation-based calibration for intractable likelihoods}},
author = {Umberto Picchini},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.39712},
eprint = {2609.39712},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.39712},
}
RIS
TY - UNPB
TI - Amortized ratio-estimation importance sampling and localized simulation-based calibration for intractable likelihoods
AU - Umberto Picchini
PY - 2026
DA - 2026/10/14
PB - arXiv
DO - 10.48550/arxiv.2609.39712
UR - https://arxiv.org/abs/2609.39712
AB - We consider simulation-based Bayesian inference (SBI) for the parameters of models with intractable likelihoods but tractable forward simulation. Building on Gaussian mixtures-of-experts surrogates, as a first contribution we develop a sequential procedure to estimate the posterior, incorporating an importance sampling step that introduces a locally amortized likelihood-to-evidence ratio estimator. The local amortization allows the ratio estimator to be evaluated at new parameter--data pairs from a local (data-dependent) design, without re-estimation, which is particularly convenient for our second contribution. The second contribution is localized simulation-based calibration (SBC). Localized SBC provides a graphical diagnostic of calibration over neighborhoods that are deliberately broader than the available posterior, at low additional computational cost. Rather than repeatedly running the full inference procedure for each simulated pseudo-observation, localized SBC fits the local surrogate and ratio estimator only once for each local design, and reuses them across all pseudo-observations. This approach is particularly appealing for SBI applications. We evaluate both contributions on three case studies, including a real-data epidemiological application.
ER -