SavedAdd your journal

Journals / Bilecik Şeyh Edebali Üniversitesi Fen Bilimleri Dergisi / 2025 / Cilt: 12 Sayı: 2

TabNet for Jet Flavor Tagging: A Machine Learning Approach

TabNet ile Jet Türü Etiketleme: Bir Makine Öğrenimi Yaklaşımı

Ali Çelik(MEHMET AKİF ERSOY ÜNİVERSİTESİ)0000-0001-8218-6512
Pages
589–599
Publication Date
2025-11-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The model achieves exceptional performance, with an accuracy of 99.98%, a recall of 99.97%, and an AUC of 1.0, demonstrating an outstanding classification performance.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Jets, collimated particle sprays from high-energy collisions, are abundant at the Large Hadron Collider and critical for studying fundamental interactions.
📊
Key Findings & Data:This study addresses the classification of hadronically decaying top quark jets and light quark or gluon jets using TabNet algorithm with leveraging low-level kinematic features of the jet constituents such as energy, transverse momentum, and components.
💡
Conclusion & Impact:The model achieves exceptional performance, with an accuracy of 99.98%, a recall of 99.97%, and an AUC of 1.0, demonstrating an outstanding classification performance.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Derin Öğrenme · Tabnet · Üst Kuark Etiketleme

Abstract

Jets, collimated particle sprays from high-energy collisions, are abundant at the Large Hadron Collider and critical for studying fundamental interactions. This study addresses the classification of hadronically decaying top quark jets and light quark or gluon jets using TabNet algorithm with leveraging low-level kinematic features of the jet constituents such as energy, transverse momentum, and components. To handle the high dimensionality of the data, Principal Component Analysis is applied for dimensionality reduction, while TabNet’s attention-based mechanism dynamically selects relevant features. The model achieves exceptional performance, with an accuracy of 99.98%, a recall of 99.97%, and an AUC of 1.0, demonstrating an outstanding classification performance.

Özet

Yüksek enerjili çarpışmalardan kaynaklanan kolimlenmiş parçacık demetleri olan jetler, Büyük Hadron Çarpıştırıcısı'nda (BHÇ) oldukça yaygındır ve temel etkileşimlerin incelenmesinde kritik bir rol oynar. Bu çalışma, hadronik olarak bozunan üst kuark jetleri ile hafif kuark veya gluon jetlerinin sınıflandırılmasını ele almakta ve bu amaçla jet bileşenlerinin enerji, enine momentum ve bileşenleri gibi düşük seviyeli kinematik özelliklerinden yararlanarak TabNet algoritmasını kullanmaktadır. Verinin yüksek boyutluluğu ile başa çıkmak için boyut indirgeme yöntemi olarak Temel Bileşenler Analizi (TBA) uygulanmakta; TabNet’in dikkat mekanizması ise dinamik olarak en ilgili özellikleri seçmektedir. Model, %99.98 doğruluk, %99.97 duyarlılık (recall) ve 1.0 AUC ile olağanüstü bir sınıflandırma performansı sergilemektedir.

🔖

Cite This Article

Çelik, A. (2025). TabNet for Jet Flavor Tagging: A Machine Learning Approach. Bilecik Şeyh Edebali Üniversitesi Fen Bilimleri Dergisi, Cilt: 12 Sayı: 2, 589–599. https://doi.org/10.35193/bseufbd.1728150

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (30)

  1. Tumasyan, A., Adam, W., Andrejkovic, J. W., Bergauer, T., Chatterjee, S., Dragicevic, M., Valle, A. E. D., Fruehwirth, R., Jeitler, M., Krammer, N., Lechner, L., Liko, D., Mikulec, I., Paulitsch, P., Pitters, F. M., Schieck, J., Schöfbeck, R., Spanring, M.,. & Vetens, W. (2022). A new calibration method for charm jet identification validated with proton-proton collision events at √s = 13 TeV. Journal of Instrumentation, 17(03), P03014. https://doi.org/10.1088/1748-0221/17/03/P03014
  2. Stoye, M. (2018). Deep learning in jet reconstruction at CMS. Journal of Physics: Conference Series, 1085, 042029. https://doi.org/10.1088/1742-6596/1085/4/042029
  3. Tumasyan, A., Adam, W., Andrejkovic, J. W., Bergauer, T., Chatterjee, S., Dragicevic, M., Escalante Del Valle, A., Frühwirth, R., Jeitler, M., Krammer, N., Lechner, L., Liko, D., Mikulec, I., Paulitsch, P., Pitters, F. M., Schieck, J., Schöfbeck, R., Schwarz, D., Templ, S., … &Vetens, W. (2022). Identification of hadronic tau lepton decays using a deep neural network. Journal of Instrumentation, 17(07), P07023. https://doi.org/10.1088/1748-0221/17/07/P07023
  4. Collaboration, A. & others. (2017). Identification of jets containing b-hadrons with recurrent neural networks at the ATLAS experiment. ATL-PHYS-PUB-2017-003.
  5. Collaboration, A. & others. (2020). Deep sets based neural networks for impact parameter flavour tagging in ATLAS. ATL-PHYS-PUB-2020-014.
  6. Sirunyan, A. M., Backhaus, M., Bäni, L., Berger, P., Bianchini, L., Dissertori, G., Dittmar, M., Donegà, M., Dorfer, C., Grab, C., …& others. (2018). Identification of heavy-flavour jets with the CMS detector in pp collisions at 13 TeV. Journal of Instrumentation, 13, P05011.
  7. Aad, G., Abbott, B., Abbott, D. C., Abud, A. A., Abeling, K., Abhayasinghe, D. K., Abidi, S. H., AbouZeid, O. S., Abraham, N. L., Abramowicz, H., Abreu, H., Abulaiti, Y., Acharya, B. S., Achkar, B., Adachi, S., Adam, L., Bourdarios, C. A., Adamczyk, L., & Zwalinski, L. (2019). ATLAS b-jet identification performance and efficiency measurement with tt ̅ events in pp collisions at √s = 13 TeV. The European Physical Journal C, 79(11), 970. https://doi.org/10.1140/epjc/s10052-019-7450-8
  8. Collaboration, C. & others. (2018). Performance of the DeepJet b tagging algorithm using 41.9 fb-1 of data from proton-proton collisions at 13 TeV with Phase 1 CMS detector. CMS Detector Performance Summary CMS-DP-2018-058.
  9. Qu, H., & Gouskos, L. (2020). Jet tagging via particle clouds. Physical Review D, 101(5), 056019. https://doi.org/10.1103/PhysRevD.101.056019
  10. Kasieczka, G., Plehn, T., Butter, A., Cranmer, K., Debnath, D., Dillon, B. M., Fairbairn, M., Faroughy, D. A., Fedorko, W., Gay, C., Gouskos, L., Kamenik, J. F., Komiske, P., Leiss, S., Lister, A., Macaluso, S., Metodiev, E., Moore, L., Nachman, B., … & Varma, S. (2019). The Machine Learning landscape of top taggers. SciPost Physics, 7(1), 014. https://doi.org/10.21468/SciPostPhys.7.1.014
  11. Arganda, E., Medina, A. D., Perez, A. D., & Szynkman, A. (2022). Towards a method to anticipate dark matter signals with deep learning at the LHC. SciPost Physics, 12(2), 063. https://doi.org/10.21468/SciPostPhys.12.2.063
  12. Baldi, P., Sadowski, P., & Whiteson, D. (2014). Searching for exotic particles in high-energy physics with deep learning. Nature Communications, 5, 4308. https://doi.org/10.1038/ncomms5308
  13. Celik, A. (2023). A fast and time-efficient machine learning approach to dark matter searches in compressed mass scenario. The European Physical Journal C, 83(12), 1150.
  14. Celik, A. (2024). Exploring hidden signal: Fine-tuning ResNet-50 for dark matter detection. Computer Physics Communications, 305, 109348. https://doi.org/10.1016/j.cpc.2024.109348
  15. Celik, A. (2024). Deep Learning Approaches for BSM Physics: Evaluating DNN and GNN Performance in Particle Collision Event Classification. Acta Physica Polonica B, 55(10), 1. https://doi.org/10.5506/APhysPolB.55.10-A2
  16. Bişkin, O. T., & Çelik, A. (2025). Transforming tabular data into graphs for GNN-driven classification of BSM and standard model events in high-energy collisions. Nuclear Physics B, 1018, 117064. https://doi.org/10.1016/j.nuclphysb.2025.117064
  17. Arik, S. Ö., & Pfister, T. (2021). TabNet: Attentive Interpretable Tabular Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 35(8), 6679–6687. https://doi.org/10.1609/aaai.v35i8.16826
  18. Almeida, L. G., Backovic, M., Cliche, M., Lee, S. J., & Perelstein, M. (2015). Playing Tag with ANN: Boosted Top Identification with Pattern Recognition (No. arXiv:1501.05968). arXiv. https://doi.org/10.48550/arXiv.1501.05968
  19. de Oliveira, L., Kagan, M., Mackey, L., Nachman, B., & Schwartzman, A. (2016). Jet-images—Deep learning edition. Journal of High Energy Physics, 2016(7), 69. https://doi.org/10.1007/JHEP07(2016)069
  20. Bols, E. S., Kieseler, J., Verzetti, M., Stoye, M., & Stakia, A. (2020). Jet Flavour Classification Using DeepJet. Journal of Instrumentation. https://doi.org/10.1088/1748-0221/15/12/p12012
  21. Furuichi, A., Lim, S. H., & Nojiri, M. M. (2024). Jet classification using high-level features from anatomy of top jets. Journal of High Energy Physics, 2024(7), 146. https://doi.org/10.1007/JHEP07(2024)146
  22. Sjöstrand, T., Ask, S., Christiansen, J. R., Corke, R., Desai, N., Ilten, P., Mrenna, S., Prestel, S., Rasmussen, C. O., & Skands, P. Z. (2015). An introduction to PYTHIA 8.2. Computer Physics Communications, 191, 159–177. https://doi.org/10.1016/j.cpc.2015.01.024
  23. de Favereau, J., Delaere, C., Demin, P., Giammanco, A., Lemaître, V., Mertens, A., Selvaggi, M., & The DELPHES 3 Collaboration. (2014). DELPHES 3: A modular framework for fast simulation of a generic collider experiment. Journal of High Energy Physics, 2014(2), 57. https://doi.org/10.1007/JHEP02(2014)057
  24. Kasieczka, G., Plehn, T., Thompson, J., & Russel, M. (2019). Top Quark Tagging Reference Dataset (Version v0 (2018_03_27)) [Dataset]. Zenodo. https://doi.org/10.5281/ZENODO.2603256
  25. McKinney, W. (2010). Data Structures for Statistical Computing in Python. 56–61. https://doi.org/10.25080/Majora-92bf1922-00a
  26. Pedregosa, F. (2011). Scikit-learn: Machine learning in python Fabian. Journal of Machine Learning Research, 12, 2825.
  27. Hotelling, H. (1933). Analysis of a complex of statistical variables into principal components. Journal of Educational Psychology, 24(6), 417–441. https://doi.org/10.1037/h0071325
  28. Pearson, K. (1901). LIII. On lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 2(11), 559–572. https://doi.org/10.1080/14786440109462720
  29. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Kopf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., … & Chintala, S. (2019). PyTorch: An Imperative Style, High-Performance Deep Learning Library. Advances in Neural Information Processing Systems, 32. https://proceedings.neurips.cc/paper/2019/hash/bdbca288fee7f92f2bfa9f7012727740-Abstract.html
  30. Tumasyan, A., Adam, W., Andrejkovic, J. W., Bergauer, T., Chatterjee, S., Dragicevic, M., Valle, A. E. D., Fruehwirth, R., Jeitler, M., Krammer, N., Lechner, L., Liko, D., Mikulec, I., Paulitsch, P., Pitters, F. M., Schieck, J., Schöfbeck, R., Spanring, M.,. & Vetens, W. (2022). A new calibration method for charm jet identification validated with proton-proton collision events at √s = 13 TeV. Journal of Instrumentation, 17(03), P03014. https://doi.org/10.1088/1748-0221/17/03/P03014