Tackling the Problem of Distributional Shifts: Correcting Misspecified, High-Dimensional Data-Driven Priors for Inverse Problems

Abstract

Bayesian inference for inverse problems hinges critically on the choice of priors. In the absence of specific prior information, population-level distributions can serve as effective priors for parameters of interest. With the advent of machine learning, the use of data-driven population-level distributions (encoded, e.g., in a trained deep neural network) as priors is emerging as an appealing alternative to simple parametric priors in a variety of inverse problems. However, in many astrophysical applications, it is often difficult or even impossible to acquire independent and identically distributed samples from the underlying data-generating process of interest to train these models. In these cases, corrupted data or a surrogate, e.g. a simulator, is often used to produce training samples, meaning that there is a risk of obtaining misspecified priors. This, in turn, can bias the inferred posteriors in ways that are difficult to quantify, which limits the potential applicability of these models in real-world scenarios. In this work, we propose addressing this issue by iteratively updating the population-level distributions by retraining the model with posterior samples from different sets of observations, and we showcase the potential of this method on the problem of background image reconstruction in strong gravitational lensing when score-based models are used as data-driven priors. We show that, starting from a misspecified prior distribution, the updated distribution becomes progressively closer to the underlying population-level distribution, and the resulting posterior samples exhibit reduced bias after several updates.

20 pages, 15 figures. To be published in The Astrophysical Journal. Added and updated references; fixed typos; extended discussions in some sections. Results unchanged

References in corpus (13)

• First M87 Event Horizon Telescope Results. II. Array and Instrumentation
• Introducing the Illustris Project: the evolution of galaxy populations across cosmic time
• astroquery: An Astronomical Web-Querying Package in Python
• Galaxy Zoo 1 : Data Release of Morphological Classifications for nearly 900,000 galaxies
• JWST Imaging of Earendel, the Extremely Magnified Star at Redshift $z=6.2$
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• IllustrisTNG in the HSC-SSP: image data release and the major role of mini mergers as drivers of asymmetry and star formation
• On the density profile of dark matter substructure in gravitational lens galaxies
• Probabilistic Mass Mapping with Neural Score Estimation
• AstroPhot: Fitting Everything Everywhere All at Once in Astronomical Images
• Event-horizon-scale Imaging of M87* under Different Assumptions via Deep Generative Image Priors
• Diffusion-HMC: Parameter Inference with Diffusion-model-driven Hamiltonian Monte Carlo
• Radio-astronomical Image Reconstruction with Conditional Denoising Diffusion Model

Cited by in corpus (2)

• Detecting Modeling Bias with Continuous Time Flow Models on Weak Lensing Maps
• Robustness of Neural Ratio and Posterior Estimators to Distributional Shifts for Population-Level Dark Matter Analysis in Strong Gravitational Lensing