Robust Amortized Simulation-Based Inference via Learned Error Models
stat.ML, astro-ph.GA, cs.LG
Submitted: 2025-09-06
Updated: 2026-11-12
Comments: Accepted to NeurIPS 2026
License: http://creativecommons.org/licenses/by/4.0/
The gist: Recent advances in neural density estimation have enabled amortized Bayesian inference for complex stochastic simulators.
Terminology
Abstract
Recent advances in neural density estimation have enabled amortized Bayesian inference for complex stochastic simulators. However, these methods rely on simulators accurately reflecting the true data-generating process and can degrade significantly under model misspecification. We consider the setting where multiple unlabeled observations are available and introduce robust variational neural posterior estimation (RVNP), an amortized Bayesian inference method that uses an importance-weighted autoencoder to jointly learn a misspecification-robust posterior proxy and an explicit error model that captures the simulation-to-reality gap. Our results show that RVNP can recover robust posterior inference in a data-driven and interpretable manner, outperforming previous methods on multiple metrics across misspecified benchmarks.
Sources
- Importance Weighted Autoencoders
- Investigating the Impact of Model Misspecification in Neural Simulation-based Inference
- Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
- Reinterpreting Importance-Weighted Autoencoders
- Simulation-Based Inference: A Practical Guide
- Neural Spline Flows
- Does Unsupervised Domain Adaptation Improve the Robustness of Amortized Bayesian Inference? A Systematic Evaluation
- Adversarial robustness of amortized Bayesian inference
- Simulation-based Bayesian inference under model misspecification
- Auto-Encoding Variational Bayes
- Amortized Variational Inference: When and Why?
- Variational Autoencoders for Efficient Simulation-Based Inference
- Masked Autoregressive Flow for Density Estimation
- Detecting Model Misspecification in Amortized Bayesian Inference with Neural Networks: An Extended Investigation
- Domain-Adaptive Neural Posterior Estimation for Strong Gravitational Lens Analysis
- Doubly Reparameterized Gradient Estimators for Monte Carlo Objectives
- Sequential Neural Posterior and Likelihood Approximation
- Simulation-Based Inference Benchmark for Weak Lensing Cosmology
Related papers
- Uniform-in-Time Weak Propagation-of-Chaos in Shallow Neural Networks
- Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
- Quantifying Error Propagation and Model Collapse in Diffusion Models
- Behavior of prediction performance metrics with rare events
- Statistical Inference and Learning for Shapley Additive Explanations (SHAP)
- Control Variate Design for Score-Function Gradient Estimators in Importance-Weighted Variational Inference