Galaxy Image Deconvolution for Weak Gravitational Lensing with Unrolled Plug-and-Play ADMM
This work addresses the challenge of improving galaxy shape measurements for cosmological studies like weak gravitational lensing, representing an incremental advance by integrating deep learning with traditional optimization techniques.
The paper tackles the problem of deblurring galaxy images for weak gravitational lensing by introducing a physics-informed deep learning method that combines algorithm unrolling and Plug-and-Play ADMM, resulting in improvements in reduced shear ellipticity error of up to 45.0% compared to classic methods and 33.2% compared to modern methods.
Removing optical and atmospheric blur from galaxy images significantly improves galaxy shape measurements for weak gravitational lensing and galaxy evolution studies. This ill-posed linear inverse problem is usually solved with deconvolution algorithms enhanced by regularisation priors or deep learning. We introduce a so-called "physics-informed deep learning" approach to the Point Spread Function (PSF) deconvolution problem in galaxy surveys. We apply algorithm unrolling and the Plug-and-Play technique to the Alternating Direction Method of Multipliers (ADMM), in which a neural network learns appropriate hyperparameters and denoising priors from simulated galaxy images. We characterise the time-performance trade-off of several methods for galaxies of differing brightness levels as well as our method's robustness to systematic PSF errors and network ablations. We show an improvement in reduced shear ellipticity error of 38.6% (SNR=20)/45.0% (SNR=200) compared to classic methods and 7.4% (SNR=20)/33.2% (SNR=200) compared to modern methods.