Sensitivity-Aware Amortized Bayesian Inference

TL;DR

This paper introduces sensitivity-aware amortized Bayesian inference (SA-ABI), a neural network-based method that efficiently incorporates sensitivity analysis into complex Bayesian models without costly refitting, enhancing interpretability and robustness.

Contribution

SA-ABI is a novel approach that combines weight sharing, rapid neural inference, and deep ensembles to perform sensitivity analysis efficiently within simulation-based Bayesian inference.

Findings

Effective in modeling disease outbreaks and climate thresholds.

Automatically detects unreliable approximations due to model misspecification.

Provides insights into hidden model sensitivities.

Abstract

Sensitivity analyses reveal the influence of various modeling choices on the outcomes of statistical analyses. While theoretically appealing, they are overwhelmingly inefficient for complex Bayesian models. In this work, we propose sensitivity-aware amortized Bayesian inference (SA-ABI), a multifaceted approach to efficiently integrate sensitivity analyses into simulation-based inference with neural networks. First, we utilize weight sharing to encode the structural similarities between alternative likelihood and prior specifications in the training process with minimal computational overhead. Second, we leverage the rapid inference of neural networks to assess sensitivity to data perturbations and preprocessing steps. In contrast to most other Bayesian approaches, both steps circumvent the costly bottleneck of refitting the model for each choice of likelihood, prior, or data set. Finally, we propose to use deep ensembles to detect sensitivity arising from unreliable approximation (e.g., due to model misspecification). We demonstrate the effectiveness of our method in applied modeling problems, ranging from disease outbreak dynamics and global warming thresholds to human decision-making. Our results support sensitivity-aware inference as a default choice for amortized Bayesian workflows, automatically providing modelers with insights into otherwise hidden dimensions.