Abstract
Liquid Argon Time Projection Chambers are used as precision detectors in ongoing and upcoming neutrino experiments. In this paper the authors adapt machine learning techniques to better identify the beginning- and end-points of particle tracks in LarTPCs, offering significant improvement over traditional methods in reconstructing the true event.
Highlights
Accelerator-based neutrino oscillation experiments have successfully deployed deep convolutional neural networks (CNNs) in their data analysis pipeline [1,2,3]
In our data set we found that 96.8% of the predicted 3D points are within 3 voxels (i.e., 0.9 cm) from the true point locations, which is promising for the future neutrino interaction vertex resolution of our full reconstruction chain
If we look at semantic-type-wise results, we find that the fraction of true points which are more than 3 voxels away from any predicted point is 7%, 2.1%, 8.2%, and 1.6% for the HIP, MIP, EM shower, and Michel electron types, respectively
Summary
Accelerator-based neutrino oscillation experiments have successfully deployed deep convolutional neural networks (CNNs) in their data analysis pipeline [1,2,3]. Many of the present and future experiments utilize a liquid argon time projection chamber (LArTPC), a class of particle imaging detectors which produce 2D or 3D images over many meters of detected charged particle trajectories, with a resolution of the order of mm/pixel. Examples of such experiments along with their respective active volumes. In the context of the DUNE, ICARUS, and SBND experiments (among others), we plan to apply it to neutrino physics analysis such as νe appearance and νμ disappearance searches
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
