Master Thesis Open Access
Schmidt, Kylian
{ "abstract": "<p>Axion-Like Particles (ALPs) are hypothetical weakly interacting light particles predicted<br>\nby theories Beyond the Standard Model which could be mediators between a dark sector<br>\nand the Standard Model. Some of these theories predict light ALPs which decay into two<br>\nphotons and could be detected at future beamdump experiments such as LUXE - New<br>\nPhysics search at Optical Dump.</p>\n\n<p><br>\nTo investigate the properties of such ALPs, an accurate reconstruction of the common<br>\nphoton vertex from the hits measured in the detector can aid the search significantly. For<br>\nthis purpose, the photon shower direction needs to be reconstructed precisely, combining<br>\ntechniques from cluster and track reconstruction. This task is a prime candidate for modern<br>\nmethods of reconstruction based on Machine Learning such as Graph Neural Networks.</p>\n\n<p><br>\nThis thesis presents a Graph Neural Network composed of GravNet and GarNet, which<br>\nis able to reconstruct the decay vertex of ALPs from the sparse detector hits of the two<br>\nphoton showers. The performance of the network is assessed on a data-set simulated with a<br>\nhigh granularity calorimeter.</p>", "author": [ { "family": "Schmidt, Kylian" } ], "id": "22245", "issued": { "date-parts": [ [ 2024, 3, 2 ] ] }, "language": "eng", "title": "Photon Reconstruction of Axion-Like Particles with Graph Neural Networks at Beamdump Experiments", "type": "thesis" }