OASIS: Observation-Aware Simulation-Based Inference via Distributional Matching

2026-06-21

AI Summary

This work addresses the mismatch between simulation outputs and real observational data in scientific modeling, which arises from measurement errors and selection effects. The authors propose explicitly embedding an observation model into the simulator and performing inference directly at the level of observed data. Their approach matches the distributions of observed and simulated data using Maximum Mean Discrepancy (MMD), integrating reweighted prior sampling with observation-aware simulation to construct a pseudo-posterior that enjoys theoretical guarantees of Monte Carlo consistency and posterior concentration. This framework eliminates the need for handcrafted summary statistics or neural surrogate models, enabling robust parameter recovery and well-calibrated uncertainty quantification in heterogeneous, non-Gaussian noise regression tasks. It has been successfully applied to complex cosmological inference from multi-band galaxy cluster observations.

Abstract

We introduce OASIS, a simulation-based inference framework for scientific settings where observations are distorted by measurement error, selection effects, and other survey-specific transformations. In many real applications, simulators generate latent, noiseless quantities, while the data are observed only after passing through a complex observational pipeline. Standard simulation-based inference methods often ignore this distinction, comparing observations to idealized simulator outputs or relying on low-dimensional summaries that can miss important structure. OASIS addresses this mismatch by explicitly embedding the observation model into the simulator and performing inference directly at the level of observed-data distributions. The method constructs a pseudo-posterior by reweighting prior samples according to a maximum mean discrepancy (MMD) loss between the empirical distributions of the observed data and forward-simulated observations, thereby avoiding both handcrafted summaries and learned neural surrogates. We provide theoretical guarantees for Monte Carlo consistency, convergence of the empirical pseudo-posterior to its population counterpart, and posterior concentration on the MMD-identified parameter set, with consistency for the true parameter under correct specification and identifiability. In controlled errors-in-variables regression experiments, OASIS delivers robust parameter recovery and well-calibrated uncertainty under heterogeneous and non-Gaussian measurement noise. We then demonstrate the method on a realistic cosmological application involving galaxy cluster observations across multiple wavelengths, in which latent physical properties are linked to observables through nonlinear scaling relations, heteroscedastic errors, selection functions, and incomplete coverage.

Problem

• simulation-based inference
• measurement error
• selection effects
• observational pipeline
• distributional mismatch

Innovation

• simulation-based inference
• observation model
• maximum mean discrepancy
• distributional matching
• measurement error