Heidelberg 2022 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 4: Deep Learning

AKPIK 4.4: Talk

Thursday, March 24, 2022, 17:00–17:15, AKPIK-H13

Binary Black Hole Parameter Reconstruction using Deep Neural Networks — •Markus Bachlechner, David Bertram, and Achim Stahl — III. Physikalisches Institut B, RWTH Aachen

The proposed Einstein Telescope, as the first of the third generation of gravitational wave detectors, is expected to be an order of magnitude more sensitive compared to current interferometers like LIGO or Virgo. On the one hand the higher sensitivity increases the observable volume. On other hand the frequency range is broadened, which in return bares the potential to extend the observable time of binary coalescences from seconds to hours. These long observable times make it possible to send multi-messenger alerts before the end of the coalescences. For this it is essential to apply a fast real-time analysis handling event detection, classification, and reconstruction. In this talk an approach for the parameter reconstruction of binary black holes using deep neural networks is presented.