Quark-Gluon Jet Discrimination Using Convolutional Neural Networks

다국어 초록 (Multilingual Abstract)

Currently, newly developed articial intelligence techniques, in particular convolutional neural networks, are being investigated for use in data-processing and classication of particle physics collider data. One such challenging task is to distinguish quark-initiated jets from gluon-initiated jets. Following previous work, we treat the jet as an image by pixelizing track information and calorimeter deposits as reconstructed by the detector. We test the deep learning paradigm by training several recently developed, state-of-the-art convolutional neural networks on the quarkgluon discrimination task. We compare the results obtained using various network architectures trained for quark-gluon discrimination and also a boosted decision tree (BDT) trained on summary variables.

참고문헌 (Reference)

1 D. Silver, 529 : 484-, 2016

2 J. Cogan, 02 : 118-, 2015

3 J. Carreira, 24 : 125-, 1998

4 M. Cacciari, C72 : 1896-, 2012

5 J. de Favereau, 1402 : 057-, 2014

6 T. Sjostrand, 191 : 159-, 2015

7 J. Alwall, 1407 : 079-, 2014

8 T. Cheng, 2 : 3-, 2018

9 M. Bronstein, 34 : 18-, 2017

10 S. Chatrchyan, 083 : S08004-, 2008

11 P. T. Komiske, 1701 : 110-, 2017

12 L. de Oliveira, 2016 : 069-, 2016

13 F. Chollet, "https://keras.io"

14 J. Hu, "arXiv:1709.01507"

15 G. Louppe, "arXiv:1702.00748"

16 G. Huang, "arXiv:1608. 06993"

17 K. He, "arXiv:1603.05027"

18 C. Szegedy, "arXiv:1602.07261"

19 K. He, "arXiv:1512.03385"

20 C. Szegedy, "arXiv:1409.4842"

21 K. Simonyan, "arXiv:1409.1556"

22 CMS Collaboration, "CMS-PAS-JME-16-003"