Abstract
The optimization of reconstruction algorithms has become a key aspect in LHCb as it is currently undergoing a major upgrade that will considerably increase the data processing rate. Aiming to accelerate the second most time consuming reconstruction process of the trigger, we propose an alternative reconstruction algorithm for the Electromagnetic Calorimeter of LHCb. Together with the use of deep learning techniques and the understanding of the current algorithm, our proposal decomposes the reconstruction process into small parts that benefit the generalized learning of small neural network architectures and simplifies the training dataset. This approach takes as input the full simulation data of the calorimeter and outputs a list of reconstructed clusters in a nearly constant time without any dependency in the event complexity.
Highlights
The increasing popularity of machine learning techniques in recent years has pushed many improvements in a wide range of computational challenges in high energy physics (HEP)
Aiming to accelerate the second most time consuming reconstruction process of the trigger, we propose an alternative reconstruction algorithm for the Electromagnetic Calorimeter of LHCb
The conclusion is to proceed with the implementation of this step using a multi layered perceptron (MLP) structure trained to be the kernel of the convolution
Summary
The increasing popularity of machine learning techniques in recent years has pushed many improvements in a wide range of computational challenges in high energy physics (HEP). Focusing in one of the main experiments, LHCb is currently facing an upgrade involving an increase in complexity and volume of data up to an average rate of 32 Tbit/s for the full 30 MHz bunch crossing rate of LHC [1] This evidences the need to improve the detectors software in efficiency and performance. We propose a reconstruction strategy for the LHCb calorimeter response that aims to reproduce the steps of the currently used algorithm in a specific formulation that facilitates the generalized learning of small deep learning structures With this methodology we achieve to have an efficient cluster reconstruction algorithm that performs at a nearly constant speed with independence of the events complexity. The results obtained in the testing of the proposal are shown, followed by a discussion and conclusions
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.
