Speaker
Description
In recent years, machine learning and deep learning have revolutionized data analysis across various fields, including particle physics and medical imaging. However, their potential in radio interferometry—a technique used to study the universe through arrays of radio telescopes—remains underexplored. The radionets-project has been pioneering the use of deep learning methods to process data from radio interferometers for over five years. These techniques promise significant improvements in performance and sensitivity, particularly for next-generation instruments like MeerKat and SKA.
Our recent publications demonstrates how deep learning can enhance imaging workflows by reconstructing high-quality images from complex datasets generated by interferometers such as VLA, VLBA, and ALMA. While initial results on simulated data have shown great promise, ongoing efforts are focused on adapting these models to real observational data from MeerKat. This poster will provide an overview of our reconstruction approach, describe the simulation pipeline we developed to train deep learning models effectively, and highlight future opportunities for applying these techniques in analyzing radio telescope data.
