Preprint
Sep 2026
Amortized ratio-estimation importance sampling and localized simulation-based calibration for intractable likelihoods
A sequential procedure to estimate the posterior is developed, incorporating an importance sampling step that introduces a locally amortized likelihood-to-evidence ratio estimator, which provides a graphical diagnostic of calibration over neighborhoods that are deliberately broader than the available posterior.
Umberto Picchini
· 0 citations