A neural network clustering algorithm for the ATLAS silicon pixel detector

G. Aad; B. Abbott; J. Abdallah; S. Abdel Khalek; O. Abdinov; R. Aben; B. Abi; Mogens Dam; Jørn Dines Hansen; Jørgen Beck Hansen; Stefania Xella; Peter Henrik Hansen; Troels Christian Petersen; Lotte Ansgaard Thomsen; Gorm Aske Gram Krohn Galster; Sascha Mehlhase; Morten Dam Jørgensen; Almut Maria Pingel; Ask Emil Løvschall-Jensen; Alejandro Alonso Diaz; James William Monk; Lars Egholm Pedersen; Graig Wiglesworth

A neural network clustering algorithm for the ATLAS silicon pixel detector

G. Aad, B. Abbott, J. Abdallah, S. Abdel Khalek, O. Abdinov, R. Aben, B. Abi, Mogens Dam, Jørn Dines Hansen, Jørgen Beck Hansen, Stefania Xella, Peter Henrik Hansen, Troels Christian Petersen, Lotte Ansgaard Thomsen, Gorm Aske Gram Krohn Galster, Sascha Mehlhase, Morten Dam Jørgensen, Almut Maria Pingel, Ask Emil Løvschall-Jensen, Alejandro Alonso DiazJames William Monk, Lars Egholm Pedersen, Graig Wiglesworth

Experimental Particle Physics

37 Citations (Scopus)

Abstract

A novel technique to identify and split clusters created by multiple charged particles in the ATLAS 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 interpolation. 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.

Original language	English
Article number	P09009
Journal	Journal of Instrumentation
Volume	9
ISSN	1748-0221
Publication status	Published - 1 Sept 2014

Access to Document

http://dx.doi.org/10.1088/1748-0221/9/09/P09009

Cite this

Aad, G., Abbott, B., Abdallah, J., Abdel Khalek, S., Abdinov, O., Aben, R., Abi, B., Dam, M., Hansen, J. D., Hansen, J. B., Xella, S., Hansen, P. H., Petersen, T. C., Thomsen, L. A., Galster, G. A. G. K., Mehlhase, S., Jørgensen, M. D., Pingel, A. M., Løvschall-Jensen, A. E., ... Wiglesworth, G. (2014). A neural network clustering algorithm for the ATLAS silicon pixel detector. Journal of Instrumentation, 9, Article P09009. http://dx.doi.org/10.1088/1748-0221/9/09/P09009

Aad, G, Abbott, B, Abdallah, J, Abdel Khalek, S, Abdinov, O, Aben, R, Abi, B, Dam, M , Hansen, JD , Hansen, JB , Xella, S , Hansen, PH , Petersen, TC, Thomsen, LA, Galster, GAGK, Mehlhase, S, Jørgensen, MD, Pingel, AM, Løvschall-Jensen, AE, Alonso Diaz, A, Monk, JW, Pedersen, LE & Wiglesworth, G 2014, 'A neural network clustering algorithm for the ATLAS silicon pixel detector', Journal of Instrumentation, vol. 9, P09009. <http://dx.doi.org/10.1088/1748-0221/9/09/P09009>

@article{bd86f342e5c345379dde52e3716bb385,

title = "A neural network clustering algorithm for the ATLAS silicon pixel detector",

abstract = "A novel technique to identify and split clusters created by multiple charged particles in the ATLAS 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 interpolation. 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.",

author = "G. Aad and B. Abbott and J. Abdallah and {Abdel Khalek}, S. and O. Abdinov and R. Aben and B. Abi and Mogens Dam and Hansen, {J{\o}rn Dines} and Hansen, {J{\o}rgen Beck} and Stefania Xella and Hansen, {Peter Henrik} and Petersen, {Troels Christian} and Thomsen, {Lotte Ansgaard} and Galster, {Gorm Aske Gram Krohn} and Sascha Mehlhase and J{\o}rgensen, {Morten Dam} and Pingel, {Almut Maria} and L{\o}vschall-Jensen, {Ask Emil} and {Alonso Diaz}, Alejandro and Monk, {James William} and Pedersen, {Lars Egholm} and Graig Wiglesworth",

year = "2014",

month = sep,

day = "1",

language = "English",

volume = "9",

journal = "Journal of Instrumentation",

issn = "1748-0221",

publisher = "Institute of Physics Publishing Ltd",

}

TY - JOUR

T1 - A neural network clustering algorithm for the ATLAS silicon pixel detector

AU - Aad, G.

AU - Abbott, B.

AU - Abdallah, J.

AU - Abdel Khalek, S.

AU - Abdinov, O.

AU - Aben, R.

AU - Abi, B.

AU - Dam, Mogens

AU - Hansen, Jørn Dines

AU - Hansen, Jørgen Beck

AU - Xella, Stefania

AU - Hansen, Peter Henrik

AU - Petersen, Troels Christian

AU - Thomsen, Lotte Ansgaard

AU - Galster, Gorm Aske Gram Krohn

AU - Mehlhase, Sascha

AU - Jørgensen, Morten Dam

AU - Pingel, Almut Maria

AU - Løvschall-Jensen, Ask Emil

AU - Alonso Diaz, Alejandro

AU - Monk, James William

AU - Pedersen, Lars Egholm

AU - Wiglesworth, Graig

PY - 2014/9/1

Y1 - 2014/9/1

N2 - A novel technique to identify and split clusters created by multiple charged particles in the ATLAS 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 interpolation. 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.

AB - A novel technique to identify and split clusters created by multiple charged particles in the ATLAS 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 interpolation. 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.

M3 - Journal article

SN - 1748-0221

VL - 9

JO - Journal of Instrumentation

JF - Journal of Instrumentation

M1 - P09009

ER -

A neural network clustering algorithm for the ATLAS silicon pixel detector

Abstract

Access to Document

Fingerprint

Cite this