A5009620496- Dark Matter and Cosmic Phenomena
- Particle Detector Development and Performance
- Radiation Detection and Scintillator Technologies
- Muon and positron interactions and applications
- Image and Signal Denoising Methods
- Neural Networks and Applications
- Radioactivity and Radon Measurements
- Atomic and Subatomic Physics Research
- Radioactive Decay and Measurement Techniques
- Particle physics theoretical and experimental studies
- Fault Detection and Control Systems
- Seismology and Earthquake Studies
- Seismic Waves and Analysis
- Seismic Imaging and Inversion Techniques
Sofia University "St. Kliment Ohridski"
2022-2025
Machine learning methods are being introduced at all stages of data reconstruction and analysis in various high-energy physics experiments. We present the development application convolutional neural networks with modified autoencoder architecture for pulse arrival time amplitude individual scintillating crystals electromagnetic calorimeters other detectors. The network performance is discussed as well xAI further investigation algorithm improvement output accuracy.
The PADME experiment at Laboratori Nationali di Frascati is designed to study possible dark matter explanations relying on the presence of a light particle that acts as mediator between visible and hidden sectors. Three data-taking runs have resulted in collection more than \(10^{13}\) positrons-on-target. experimental setup was initially search for associate photon production later modified resonant new \(X17\) particle, originally proposed by ATOMKI Institute. different runs, data taking,...
Abstract Machine learning methods find growing application in the reconstruction and analysis of data high energy physics experiments. A modified convolutional autoencoder model was employed to identify reconstruct pulses from scintillating crystals. The further investigated using four xAI for deeper understanding underlying mechanism. results are discussed detail, underlining importance knowledge gain improvement algorithms.
Abstract The PADME Experiment at the Laboratori Nationali di Frascati, INFN is used in search for a Dark photon, produced with an ordinary photon electron-positron annihilation events. energy of photons, emitted measured using segmented electromagnetic calorimeter. Machine learning methods consisting various convolutional neural networks are reconstruction close-in-time signals high resolution. These algorithms were on two-photon events e + − → γγ to calibrate values. In order network output...
The PADME apparatus was built at the Frascati National Laboratory of INFN to search for a dark photon ($A'$) produced via process $e^+ e^- \rightarrow A' \gamma$. central component detector is an electromagnetic calorimeter composed 616 BGO crystals dedicated measurement energy and position final state photons. high beam particle multiplicity over short bunch duration requires reliable identification overlapping signals. A regression machine-learning-based algorithm has been developed...
A bstract This paper presents a detailed characterization of the positron beam delivered by Beam Test Facility at Laboratori Nazionali Frascati to PADME experiment during Run III, which took place from October December 2022. It showcases methodology used measure main parameters such as position in space, absolute momentum scale, energy spread, and its intensity through combination data analysis Monte Carlo simulations. The results achieved include an precision within ~1–2 MeV /c , relative...
Abstract Machine learning methods can be used for signal processing in different cases of physics research. A convolutional neural network was developed the task pulse counting particle detectors high energy physics. For extraction parameters a with autoencoder architecture and subsequent result reconstruction algorithm applied. also seismic studies identifying events seismograms. All algorithms their architecture, input output are presented discussed.