|
Estimation of Gravitational Lens Parameters at Intermediate Redshifts Using Convolutional Neural Networks (CNN) a) Faculty of Mathematics and Natural Sciences, Bandung Institute of Technology Abstract Strong gravitational lensing serves as a powerful astrophysical probe, enabling studies of dark matter, galaxy structure, and cosmological parameters. The number of strong gravitational lensing candidates at the galaxy scale is expected to reach O~10^{5} with ongoing and future wide-field galaxy surveys. Current modeling techniques largely depend on conventional fitting-such as least squares or maximum likelihood using Markov Chain Monte Carlo (MCMC)-which, despite their effectiveness, are computationally expensive and demand manual inspection. This motivates the development of faster yet accurate parameter estimation techniques. In this work, we construct a representative training dataset and develop an efficient Convolutional Neural Network (CNN) to estimate crucial lens parameters: the Einstein radius, axis ratio, and position angle. We utilize data from Public Data Release 3 (PDR3) of the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). Lens galaxies are selected in the range 0.3<z<0.9 following the strong-lens probability distribution. Preliminary results indicate that both the selection of the loss function and the regularization strategy markedly influence model performance. SpatialDropout outperforms standard dropout for regularization. Furthermore, prediction accuracy and convergence speed depend heavily on the distribution of the training data, thereby informing the optimal choice of loss function. Keywords: Strong gravitational lensing, Convolutional Neural Networkc (CNN), Lens parameter estimation Topic: Instrumentation and Computational Physics |
| IPS 2025 Conference | Conference Management System |