Speaker
Kenny Nguyen
(University of Waterloo)
Description
The Event Horizon Telescope (EHT) is a very long baseline interferometry (VLBI) array that has the capacity to resolve images of supermassive black holes such as Sagittarius A and M87. Turbulence in the interstellar medium distorts images of objects near the galactic center, e.g., Sagittarius A*. This reduces the angular resolution that could be resolved. The scattering screen changes on time scales that are longer than the scales that EHT uses. We utilized a recurrent neural network to mitigate these effects. The model resolves multiple images concurrently by using the long-time scale property of interstellar scattering. We used training samples that are agnostic to General Relativity.
| Keyword-1 | Neural networks |
|---|---|
| Keyword-2 | Interstellar scattering |
| Keyword-3 | Supermassive black holes |
Author
Kenny Nguyen
(University of Waterloo)
Co-authors
Arvin Kouroshnia
Dr
Chunchong Ni
(University of Waterloo)
Mr
Ali SaraerToosi
(University of Toronto)
Dr
Avery Broderick
(University of Waterloo)
