High-dimensional inference for the $γ$-ray sky with differentiable programming
2026-04-09 • Machine Learning
Machine Learning
AI summaryⓘ
The authors use a special type of programming called differentiable probabilistic programming to study the complicated space of models used to analyze gamma-ray data from space. They focus on the mysterious excess of gamma rays coming from the center of our galaxy and build tools that use powerful graphics processors to quickly test many possible explanations at once. Their approach helps to make better guesses about what causes this excess by considering many shapes and patterns in the data probabilistically. They also want to show that this method can be useful for other astrophysics studies with large datasets.
differentiable probabilistic programminggamma raysGalactic Center gamma-ray Excessforward modelinglikelihood functionGPU accelerationvectorizationvariational inferencespatial morphologyastrophysical data analysis
Authors
Siddharth Mishra-Sharma, Tracy R. Slatyer, Yitian Sun, Yuqing Wu
Abstract
We motivate the use of differentiable probabilistic programming techniques in order to account for the large model-space inherent to astrophysical $γ$-ray analyses. Targeting the longstanding Galactic Center $γ$-ray Excess (GCE) puzzle, we construct differentiable forward model and likelihood that make liberal use of GPU acceleration and vectorization in order to simultaneously account for a continuum of possible spatial morphologies consistent with the GCE emission in a fully probabilistic manner. Our setup allows for efficient inference over the large model space using variational methods. Beyond application to $γ$-ray data, a goal of this work is to showcase how differentiable probabilistic programming can be used as a tool to enable flexible analyses of astrophysical datasets.