Speaker
Description
Alpha Magnetic Spectrometer (AMS-02) is a precision high-energy cosmic-ray experiment consisting of Transition Radiation Detector (TRD), Silicon Tracker, Magnet, Time of Flight (ToF), and Ring Imaging Cherenkov Detector (RICH), Anti-Coincidence Counter (ACC), and Electromagnetic Calorimeter (ECAL) on the ISS operating since 2011, and has collected more than 240 billion cosmic-ray events. Among them, positrons are important in understanding the particle nature of dark matter if it annihilates into positrons. Separating the positron signals from the abundant background protons is challenging because the protons can also produce showers with an electromagnetic component in the ECAL via neutral pion decays, mimicking electron/positron showers. Therefore, we use a state-of-the-art convolutional and transformer model, such as CoAtNet, that employs all energy depositions in the ECAL cells, represented in matrix format as images, to classify the electrons/positrons in the dominant cosmic proton background. We created training, validation, and test datasets for CoAtNet, in which purer samples of electrons and protons from ISS data and test beam data in the 50-200 GeV reconstructed energy range are selected by using information from Tracker, ToF, and TRD. To reject charge confused protons from the electron sample and create datasets in the 200-500 GeV range, we built an additional charge-confused proton and electron classifier with a neural network using Tracker, ToF, and RICH information. We trained and fine-tuned the ECAL CoAtNet using the datasets and obtained a proton rejection that outperforms the traditional methods by more than a factor of 2 on ISS Data in the 50-500 GeV range.
