Speakers
Description
The Early Data Release and eRASS1 data from the eROSITA space telescope have already revealed a remarkable number of previously undetected X-ray sources. Leveraging Bayesian inference and generative modeling techniques for X-ray imaging, we aim to enhance the sensitivity and scientific value of these observations by denoising, deconvolving, and decomposing the X-ray sky. Utilizing information field theory, we exploit the spatial and spectral correlation structures of various sky components with non-parametric prior models to improve their reconstruction.
By incorporating the instrument's point-spread function, exposure, and effective area information from the calibration database into our forward model, we seek to develop a comprehensive Bayesian imaging algorithm for the eROSITA Western Galactic Hemisphere. This approach aims to enhance the existing X-ray source catalogs therefore advancing our understanding of the X-ray universe.
