Die DPG-Frühjahrstagung in Bonn musste abgesagt werden! Lesen Sie mehr ...

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 1: AKPIK I

AKPIK 1.6: Vortrag

Mittwoch, 1. April 2020, 17:45–18:00, H-HS XII

Analyzing Radio Interferometric Data with Neural Networks — •Kevin Schmidt, Felix Geyer, and Simone Mender — TU Dortmund

Very long baseline radio interferometry allows the observation of distant astronomical objects with the highest resolution. In this technique, the data of several radio telescopes are combined to achieve an effective diameter equal to the greatest baseline.

Radio interferometers measure visibilities depending on the baselines between the individual telescopes. Based on their sparse distribution, much visibility space remains uncovered. This lack of information causes noise artifacts in the recorded data, which have to be removed to receive a clean image.

With increasing data rates of modern radio interferometers, fast solutions are necessary to analyze observations in a reasonable time. One approach is the usage of machine learning techniques like neural networks. In this talk, the feasibility study of reconstructing sparse radio interferometric data using convolutional neural networks is presented based on a toy Monte Carlo data set.