Abstract: The XENON1T experiment uses a time projection chamber (TPC) with liquid xenon to search for weakly interacting massive particles (WIMPs), a proposed dark matter particle, via direct detection. As this experiment relies on capturing rare events, the focus is on achieving a high recall of WIMP events. Hence the ability to distinguish between WIMP and the background is extremely important. To accomplish this, we suggest using convolutional neural networks (CNNs); a machine learning procedure mainly used in image recognition tasks. To explore this technique we use XENON collaboration open-source software to simulate the TPC graphical output of dark matter signals and main backgrounds. A CNN turns out to be a suitable tool for this purpose, as it can identify features in the images that differentiate the two types of events without the need to manipulate or remove data in order to focus on a particular region of the detector. We find that the CNN can distinguish between the dominant background events (ER) and 500 GeV WIMP events with a recall of 93.4%, precision of 81.2% and an accuracy of 87.2%.

Abstract: We report the performance of a 10 atm Xenon/trimethylamine time projection chamber (TPC) for the detection of X-rays (30 keV) and gamma-rays (0.511-1.275 MeV) in conjunction with the accurate tracking of the associated electrons. When operated at such a high pressure and in similar to 1%-admixtures, trimethylamine (TMA) endows Xenon with an extremely low electron diffusion (1.3 +/- 0.13 mm-sigma (longitudinal), 0.95 +/- 0.20 mm-sigma (transverse) along 1 m drift) besides forming a convenient Penning-Fluorescent' mixture. The TPC, that houses 1.1 kg of gas in its fiducial volume, operated continuously for 100 live-days in charge amplification mode. The readout was performed through the recently introduced microbulk Micromegas technology and the AFTER chip, providing a 3D voxelization of 8 mm x 8 mm x 1.2 mm for approximately 10 cm/MeV-long electron tracks. Resolution in energy (epsilon) at full width half maximum (R) inside the fiducial volume ranged from R = 14.6% (30 keV) to R = 4.6% (1.275 MeV). This work was developed as part of the R&D program of the NEXT collaboration for future detector upgrades in the search of the neutrino-less double beta decay (beta beta 0 nu) in Xe-136, specifically those based on novel gas mixtures. Therefore we ultimately focus on the calorimetric and topological properties of the reconstructed MeV-electron tracks. In particular, the obtained energy resolution has been decomposed in its various contributions and improvements towards achieving the R =1.4%root MeV/epsilon levels obtained in small sensors are discussed.