A neural network clustering algorithm for the ATLAS silicon pixel detector

Aad, G. and Abbott, B. and Abdallah, J. and Khalek, S. A. and Abdinov, O. and Aben, R. and Abi, B. and Abolins, M. and AbouZeid, O. S. and Abramowicz, H. and Abreu, H. and Abreu, R. and Abulaitia, Y. and Acharya, B. S. and Adamczyka, L. and Adams, D. L. and Adelman, J. and Adomeit, S. and Adye, T. and Agatonovic-Jovin, T. and Aguilar-Saavedra, J. A. and Agustoni, M. and Ahlen, S. P. and Ahmadov, F. and Aielli, G. and Akerstedt, H. and Akesson, T. P. A. and Akimoto, G. and Akimov, A. V. and Alberghi, G. L. and Krasznahorkay, Attila and Nagai, Yoshikazu and Pásztor, Gabriella and Tóth, József (2014) A neural network clustering algorithm for the ATLAS silicon pixel detector. JOURNAL OF INSTRUMENTATION, 9. ISSN 1748-0221

Preview

Text 1406.7690.pdf Available under License Creative Commons Attribution. Download (1MB) | Preview

Abstract

A novel technique to identify and split clusters created by multiple charged particles in the AT- LAS pixel detector using a set of artificial neural networks is presented. Such merged clusters are a common feature of tracks originating from highly energetic objects, such as jets. Neural networks are trained using Monte Carlo samples produced with a detailed detector simulation. This technique replaces the former clustering approach based on a connected component analysis and charge in- terpolation. The performance of the neural network splitting technique is quantified using data from proton–proton collisions at the LHC collected by the ATLAS detector in 2011 and from Monte Carlo simulations. This technique reduces the number of clusters shared between tracks in highly energetic jets by up to a factor of three. It also provides more precise position and error estimates of the clusters in both the transverse and longitudinal impact parameter resolution.

Actions (login required)

Edit Item