A data-driven event generator for Hadron Colliders using Wasserstein Generative Adversarial Network

Abstract

Highly reliable Monte-Carlo event generators and detector simulation programs are important for the precision measurement in the high energy physics. Huge amounts of computing resources are required to produce a sufficient number of simulated events. Moreover, simulation parameters have to be fine-tuned to reproduce situations in the high-energy particle interactions which is not trivial in some phase spaces in physics interests. In this paper, we suggest a new method based on the Wasserstein Generative Adversarial Network (WGAN) that can learn the probability distribution of the real data. Our method is capable of event generation at a very short computing time compared to the traditional MC generators. The trained WGAN is able to reproduce the shape of the real data with high fidelity.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

References

  1. 1.

    I. Goodfellow et al., Proceedings of the International Conference on Neural Information Processing Systems (NIPS 2014), pp. 2672–2680 (2014)

  2. 2.

    M. Paganini, L. Oliveira, B. Nachman, Phys. Rev. D 97, 014021 (2018)

    ADS  Article  Google Scholar 

  3. 3.

    M. Paganini, L. Oliveira, B. Nachman, Phys. Rev. Lett. 120, 042003 (2018)

    ADS  Article  Google Scholar 

  4. 4.

    B. Hashemi, N. Amin, K. Datta, D. Olivito, M. Pierini,. arXiv:1901.05282 [hep-ex]

  5. 5.

    R.D. Sipio, M.F. Giannelli, S.K. Haghighat, S. Palazzo, J. High Energy Phys. 08, 110 (2019)

    ADS  Article  Google Scholar 

  6. 6.

    M. Arjovsky, S. Chintala, L. Bottou, ICML’17: Proceedings of the 34th International Conference on Machine Learning, 70 (ICML 2017) pp. 214–223 (2017)

  7. 7.

    I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin, A. Courville, NIPS’17: Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS 2017) pp. 5769–5779 (2017)

  8. 8.

    M. Erdmann, L. Geiger, J. Glombitza, D. Schmidt, Comput. Softw. Big Sci. 2, 4 (2018)

    Article  Google Scholar 

  9. 9.

    N. Rahaman et al., Proceedings of the 36th International Conference on Machine Learning (ICML 2019) pp. 5301–5310 (2019)

  10. 10.

    J. Alwall, M. Herquet, F. Maltoni, O. Mattelaer, T. Stelzer, J. High Energy Phys. 06, 128 (2011)

    ADS  Article  Google Scholar 

  11. 11.

    T. Sjstrand et al., Comput. Phys. Commun. 191, 159–177 (2015)

    ADS  Article  Google Scholar 

  12. 12.

    J. de Favereau, C. Delaere, P. Demin, A. Giammanco, V. Lematre, A. Mertens, M. Selvaggi, J. High Energy Phys. 02, 057 (2014)

    Article  Google Scholar 

  13. 13.

    The CMS Collaboration, J. Instrum. 3, 8004 (2008)

    Article  Google Scholar 

  14. 14.

    M. Cacciari, G.P. Salam, G. Soyez, J. High Energy Phys. 04, 063 (2008)

    ADS  Article  Google Scholar 

  15. 15.

    I. Durugkar, I. Gemp, S. Mahadevan, Conference paper of International Conference on Learning Representations (ICLR 2017) (2017)

  16. 16.

    M. Abadi et al., https://www.tensorflow.org (2015)

  17. 17.

    F. Chollet et al., https://keras.io (2015)

  18. 18.

    L. Mason, J. Baxter, P. Bartlett, M. Frean, Advances in Neural Information Processing Systems 12 (MIT Press, New York, 1999), pp. 512–518

    Google Scholar 

Download references

Acknowledgements

This work is supported by the National Research Foundation of Korea (NRF) under Contract No. NRF-2018R1A2B6005043, NRF-2020R1A2C3009918, and the BK21 FOUR program at Korea University, Initiative for science frontiers on upcoming challenges.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Jae Hoon Lim.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

A. Normalized distributions of momentum components

See Figs. 5, 6 and 7.

Fig. 5
figure5

The distributions of three momentum components \((p_x, p_y, p_z)\) of the subleading photon

Fig. 6
figure6

The distributions of three momentum components \((p_x, p_y, p_z)\) of the leading b-jet

Fig. 7
figure7

The distributions of three momentum components \((p_x, p_y, p_z)\) of the subleading b-jet

B. Normalized distributions of additional variables

See Figs. 8, 9, 10 and 11.

Fig. 8
figure8

The distributions of \(p_{\text {T}}\) and \(\eta\) of the subleading photon

Fig. 9
figure9

The distributions of \(p_{\text {T}}\) and \(\eta\) of the leading b-jet

Fig. 10
figure10

The distributions of \(p_{\text {T}}\) and \(\eta\) of the subleading b-jet

Fig. 11
figure11

The distributions of invariant mass of two b-jets and \(\Delta R (b\text {-jet}_1, b\text {-jet}_2\))

C. Normalized distributions of the control region

See Fig. 12.

Fig. 12
figure12

The distributions of the invariant mass of two photons and \(\Delta R (\gamma _{1}, \gamma _{2})\) with the condition of [100, 150] GeV

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Choi, S., Lim, J.H. A data-driven event generator for Hadron Colliders using Wasserstein Generative Adversarial Network. J. Korean Phys. Soc. 78, 482–489 (2021). https://doi.org/10.1007/s40042-021-00095-1

Download citation

Keywords

  • HEP data
  • Event generation
  • Deep learning
  • GAN
  • WGAN