Modeling strong gravitational lenses in order to quantify the distortions of the background sources and reconstruct the mass density in the foreground lens has traditionally been a major computational challenge. As the quality of gravitational lens images increases, the task of fully exploiting the infor- mation they contain becomes computationally and algorithmically more difficult. In this work, we use a neural network based on the Recurrent Inference Machine (RIM) to simultaneously reconstruct an undistorted image of the background source and the lens mass density distribution as pixelated maps. The method we present iteratively reconstructs the model parameters (the source and density map pixels) by learning the process of optimization of their likelihood given the data using the physical model (a ray tracing simulation), regularized by a prior implicitly learnt by the neural network through its training data. When compared to more traditional parametric models, the method we propose is significantly more expressive and can reconstruct complex mass distribution, which we demonstrate by using realistic lensing galaxies taken from the hydrodynamical IllustrisTNG simulation .
Alexandre Adam |
Laurence Perreault-Levasseur |
Yashar Hezaveh |