A5015258628- Particle Detector Development and Performance
- Particle physics theoretical and experimental studies
- Radiation Detection and Scintillator Technologies
- High-Energy Particle Collisions Research
- Nuclear Physics and Applications
- Machine Learning and Data Classification
- Advanced Multi-Objective Optimization Algorithms
- Experimental and Theoretical Physics Studies
- Atomic and Subatomic Physics Research
- Astronomical Observations and Instrumentation
- Parallel Computing and Optimization Techniques
- Optical measurement and interference techniques
- Evolutionary Algorithms and Applications
Joint Institute for Nuclear Research
2021-2024
Dubna State University
2022
St Petersburg University
2022
The reconstruction of charged particle trajectories in tracking detectors is a key problem the analysis experimental data for high-energy and nuclear physics. amount modern experiments so large that classical methods, such as Kalman filter, cannot process them fast enough. To solve this problem, we developed two neural network algorithms based on deep learning architectures track recognition pixel strip-based detectors. These are TrackNETv3 local (track by track) RDGraphNet global (all...
Tracking is an important task in the field of High Energy physics. Modern experiments produceenormous amounts data, and classical tracking algorithms cannot reach required computingefficiency. This lead to need develop new methods, some them use neural network models.In our work we present modifications previously developed model, TrackNetV2. model itsdescendants showed great results for Monte-Carlo simulations with microstrip-basedGEM detectors: BESIII BM@N RUN6. In this adapt it more...
Particle tracking is an essential part of any high-energy physics experiment. Well-known algorithms based on the Kalman filter are not scaling well with amounts data being produced in modern experiments. In our work we present a particle approach deep neural networks for BM@N experiment and future SPD We have already applied similar approaches RUN 6 BES-III Monte-Carlo simulation data. This next step ongoing study help machine learning. Revised - combination Recurrent Neural Network (RNN)...
Modern machine learning (ML) tasks and neural network (NN) architectures require huge amounts ofGPU computational facilities demand high CPU parallelization for data preprocessing. At thesame time, the Ariadne library, which aims to solve complex high-energy physics tracking withthe help of deep networks, lacks multi-GPU training efficient parallel preprocessing onthe CPU.In our work, we present approach Multi-GPU in library. We willpresent data-caching, preprocessing, generic ML experiment...