Beschreibung
A fundamental question regarding the Fermi gamma-ray Galactic Center Excess (GCE) is whether its underlying structure is point-like or smooth. This debate, often framed in terms of a millisecond pulsar or annihilating dark matter origin for the emission, awaits a conclusive resolution.I will present how deep learning tools can be used to distinguish between these two possibilities. Specifically, we weigh in on the problem using graph-convolutional neural networks and estimate the fluxes of different emission components as well as the source-count distributions of the underlying point-source populations. In the Fermi data, we find a GCE composed of a large number of faint sources, with N >∼ O(10,000) sources required to explain the entire excess (median value N = 29,300 across the sky). This could be accommodated by recent millisecond pulsar population studies, but in view of the inherent degeneracy between faint point-sources and Poissonian ("dark-matter like") emission, the presence of a GCE component due to dark matter annihilation remains viable.
| Zeitraum | 27 Juni 2022 |
|---|---|
| Gehalten am | Universität Bremen, Deutschland |