Christopher Danner
I am a UWA grad student working with Richard Dodson on a PhD in astronomy. My research is focused on applying machine learning methods to improve radio interferometry imaging data.
Session
We are on the cusp of astronomical analysis becoming data-swamped, with several next-generation instruments (SKA, Next Generation Very Large Array (ngVLA), Next Generation Event Horizon Telescope (ngEHT), etc.) related to radio-interferometry that are either in their final stages of planning or already under commission. The current method for cleaning dirty radio images is slow and difficult to automate, making it extremely challenging to handle large quantities of data. This will create a bottleneck in the data pipeline when the next-generation interferometers are finished being built. In order to not limit the discoveries that will come from these next-generation instruments, a more efficient method for processing data needs to be created. In this project, we use synthetic radio interferometry data to train a machine learning algorithm capable of deconvolving (‘cleaning’) dirty radio images. The goal of this project is to develop an automated framework that is faster and more accurate compared to current image cleaning techniques.