arXiv:2605.22112v1 Announce Type: cross Abstract: We present a framework for detecting transient gamma-ray phenomena in a controlled environment by combining end-to-end simulations of the Fermi-LAT sky with self-supervised spatio-temporal deep learning. We generate a ten-year synthetic Universe with gtobssim and process the simulated events into daily all-sky maps of counts and exposure, obtaining a time-ordered sequence that mirrors the structure of Fermi-LAT observations. To model the nominal evolution of the sky, we employ a Convolutional Long Short-Term Memory (ConvLSTM) network that opera
The continuous advancements in AI and deep learning are enabling the application of sophisticated models like ConvLSTM to complex scientific data analysis, aligning with current trends in AI-driven scientific discovery.
This development showcases the growing capability of AI to automate and enhance the detection of transient astronomical phenomena, leading to faster and more accurate scientific insights.
The adoption of self-supervised ConvLSTM could significantly improve the efficiency and sensitivity of transient detection in high-energy astrophysics, potentially revealing previously unobserved events.
- · Astrophysicists
- · Space agencies
- · AI researchers
- · High-energy astronomy community
- · Manual data analysis methods
More efficient detection of gamma-ray transients in astronomical data.
Accelerated understanding of extreme astrophysical events and fundamental physics.
New discoveries in high-energy astrophysics enabled by AI-driven analysis, potentially leading to novel theoretical models.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG