Scalable Bayesian Inference for Finding Strong Gravitational Lenses

TL;DR

This paper introduces a scalable Bayesian inference method for detecting strong gravitational lenses in astronomical images, providing well-calibrated uncertainties and overcoming limitations of previous approaches.

Contribution

It presents a fully amortized Bayesian procedure trained with forward KL divergence, enabling scalable and uncertainty-aware lens detection in large surveys.

Findings

Effective in synthetic GalSim images

Accurate uncertainty quantification in SDSS images

Outperforms traditional variational inference methods

Abstract

Finding strong gravitational lenses in astronomical images allows us to assess cosmological theories and understand the large-scale structure of the universe. Previous works on lens detection do not quantify uncertainties in lens parameter estimates or scale to modern surveys. We present a fully amortized Bayesian procedure for lens detection that overcomes these limitations. Unlike traditional variational inference, in which training minimizes the reverse Kullback-Leibler (KL) divergence, our method is trained with an expected forward KL divergence. Using synthetic GalSim images and real Sloan Digital Sky Survey (SDSS) images, we demonstrate that amortized inference trained with the forward KL produces well-calibrated uncertainties in both lens detection and parameter estimation.