Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

ν-Flows: Conditional Neutrino Regression (2207.00664v7)

Published 1 Jul 2022 in hep-ph, cs.LG, and hep-ex

Abstract: We present $\nu$-Flows, a novel method for restricting the likelihood space of neutrino kinematics in high energy collider experiments using conditional normalizing flows and deep invertible neural networks. This method allows the recovery of the full neutrino momentum which is usually left as a free parameter and permits one to sample neutrino values under a learned conditional likelihood given event observations. We demonstrate the success of $\nu$-Flows in a case study by applying it to simulated semileptonic $t\bar{t}$ events and show that it can lead to more accurate momentum reconstruction, particularly of the longitudinal coordinate. We also show that this has direct benefits in a downstream task of jet association, leading to an improvement of up to a factor of 1.41 compared to conventional methods.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (64)
  1. L. R. Evans and P. Bryant, LHC Machine, JINST 3, S08001 (2008), 10.1088/1748-0221/3/08/S08001.
  2. ATLAS Collaboration, The ATLAS experiment at the CERN Large Hadron Collider, JINST 3, S08003 (2008), 10.1088/1748-0221/3/08/S08003.
  3. CMS Collaboration, The CMS experiment at the CERN LHC, JINST 3, S08004 (2008), 10.1088/1748-0221/3/08/S08004.
  4. ATLAS Collaboration, Identification of electrons using a deep neural network in the ATLAS experiment, ATL-PHYS-PUB-2022-022 (2022).
  5. ATLAS Collaboration, ATLAS b-jet identification performance and efficiency measurement with t⁢t¯𝑡normal-¯𝑡t{\bar{t}}italic_t over¯ start_ARG italic_t end_ARG events in pp collisions at s=13𝑠13\sqrt{s}=13square-root start_ARG italic_s end_ARG = 13 TeV, Eur. Phys. J. C 79, 970 (2019), 10.1140/epjc/s10052-019-7450-8.
  6. ATLAS Collaboration, Graph Neural Network Jet Flavour Tagging with the ATLAS Detector, ATL-PHYS-PUB-2022-027 (2022).
  7. M. Stoye (on behalf of the CMS Collaboration), Deep learning in jet reconstruction at CMS, J. Phys.: Conf. Ser. 1085, 042029 (2018), 10.1088/1742-6596/1085/4/042029.
  8. G. Degrassi et al., Higgs mass and vacuum stability in the standard model at NNLO, JHEP 08, 98 (2012), 10.1007/jhep08(2012)098.
  9. E. G. Tabak and E. Vanden-Eijnden, Density estimation by dual ascent of the log-likelihood, Commun. Math. Sci. 8(1), 217 (2010), 10.4310/CMS.2010.v8.n1.a11.
  10. D. Rezende and S. Mohamed, Variational inference with normalizing flows, In Proceedings of the 32nd International Conference on Machine Learning, vol. 37 of PMLR, pp. 1530–1538. Lille, France (2015), 1505.05770.
  11. CMS Collaboration, Measurement of differential cross sections for top quark pair production using the lepton+jets final state in proton-proton collisions at 13 TeV, Phys. Rev. D 95, 092001 (2017), 10.1103/PhysRevD.95.092001.
  12. CMS Collaboration, Measurement of differential cross sections for the production of top quark pairs and of additional jets in lepton+jets events from pp collisions at s=𝑠absent\sqrt{s}=square-root start_ARG italic_s end_ARG = 13 TeV, Phys. Rev. D 97, 112003 (2018), 10.1103/PhysRevD.97.112003.
  13. ATLAS Collaboration, Measurements of top-quark pair differential and double-differential cross-sections in the ℓnormal-ℓ\ellroman_ℓ+jets channel with p⁢p𝑝𝑝ppitalic_p italic_p collisions at s=13𝑠13\sqrt{s}=13square-root start_ARG italic_s end_ARG = 13 TeV using the ATLAS detector, Eur. Phys. J. C 79, 1028 (2019), 10.1140/epjc/s10052-019-7525-6, [Erratum: Eur. Phys. J. C 80, 1092 (2020)].
  14. CMS Collaboration, Measurement of differential t⁢t¯𝑡normal-¯𝑡t\bar{t}italic_t over¯ start_ARG italic_t end_ARG production cross sections in the full kinematic range using lepton+jets events from proton-proton collisions at s𝑠\sqrt{s}square-root start_ARG italic_s end_ARG = 13  TeV, Phys. Rev. D 104, 092013 (2021), 10.1103/PhysRevD.104.092013.
  15. ATLAS Collaboration, Determination of the top-quark pole mass using t⁢t¯𝑡normal-¯𝑡t\overline{t}italic_t over¯ start_ARG italic_t end_ARG + 1-jet events collected with the ATLAS experiment in 7 TeV pp collisions, JHEP 10, 121 (2015), 10.1007/JHEP10(2015)121.
  16. CMS Collaboration, Measurement of the top quark mass with lepton+jets final states using p⁢p𝑝𝑝ppitalic_p italic_p collisions at s=13⁢𝑇𝑒𝑉𝑠13𝑇𝑒𝑉\sqrt{s}=13\,\text{TeV}square-root start_ARG italic_s end_ARG = 13 TeV, Eur. Phys. J. C 78, 891 (2018), 10.1140/epjc/s10052-018-6332-9, [Erratum: Eur. Phys. J. C 82, 323 (2022)].
  17. ATLAS Collaboration, Measurement of the top quark mass in the t⁢t¯→normal-→𝑡normal-¯𝑡absentt\bar{t}\rightarrowitalic_t over¯ start_ARG italic_t end_ARG → lepton+jets channel from s=8𝑠8\sqrt{s}=8square-root start_ARG italic_s end_ARG = 8 TeV ATLAS data and combination with previous results, Eur. Phys. J. C 79, 290 (2019), 10.1140/epjc/s10052-019-6757-9.
  18. ATLAS Collaboration, Measurement of the top-quark mass in t⁢t¯+1𝑡normal-¯𝑡1t\bar{t}+1italic_t over¯ start_ARG italic_t end_ARG + 1-jet events collected with the ATLAS detector in p⁢p𝑝𝑝ppitalic_p italic_p collisions at s=8𝑠8\sqrt{s}=8square-root start_ARG italic_s end_ARG = 8 TeV, JHEP 11, 150 (2019), 10.1007/JHEP11(2019)150.
  19. K. Zoch et al., Semileptonic ttbar neutrino regression dataset, 10.5281/zenodo.6782987 (2022).
  20. C. Durkan et al., Neural spline flows, In Proceedings of Advances in Neural Information Processing Systems, vol. 32. Vancouver, Canada (2019), 1906.04032.
  21. D. P. Kingma and P. Dhariwal, Glow: Generative flow with invertible 1x1 convolutions, In Proceedings of Advances in Neural Information Processing Systems, vol. 31. Montreal, Canada (2018), 1807.03039.
  22. B. Ross and J. Cresswell, Tractable density estimation on learned manifolds with conformal embedding flows, In Proceedings of Advances in Neural Information Processing Systems, vol. 34, pp. 26635–26648 (2021), 2106.05275.
  23. L. Ardizzone et al., Guided image generation with conditional invertible neural networks (2019), 1907.02392.
  24. B. Stienen and R. Verheyen, Phase Space Sampling and Inference from Weighted Events with Autoregressive Flows, SciPost Phys. 10(2), 38 (2021), 10.21468/scipostphys.10.2.038.
  25. B. Nachman and D. Shih, Anomaly detection with density estimation, Phys. Rev. D 101, 075042 (2020), 10.1103/PhysRevD.101.075042.
  26. A. Hallin et al., Classifying anomalies through outer density estimation, Phys. Rev. D 106, 055006 (2022), 10.1103/PhysRevD.106.055006.
  27. J. A. Raine et al., Curtains for your sliding window: Constructing unobserved regions by transforming adjacent intervals (2022), 2203.09470.
  28. J. Brehmer and K. Cranmer, Flows for simultaneous manifold learning and density estimation, In Proceedings of Advances in Neural Information Processing Systems, vol. 33, pp. 442–453 (2020), 2003.13913.
  29. M. Bellagente et al., Invertible Networks or Partons to Detector and Back Again, SciPost Phys. 9(5), 74 (2020), 10.21468/scipostphys.9.5.074.
  30. C. Krause and D. Shih, CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows (2021), 2106.05285.
  31. C. Krause and D. Shih, CaloFlow II: Even Faster and Still Accurate Generation of Calorimeter Showers with Normalizing Flows (2021), 2110.11377.
  32. M. Grossi et al., Comparing traditional and deep-learning techniques of kinematic reconstruction for polarization discrimination in vector boson scattering, Eur. Phys. J. C 80, 1144 (2020), 10.1140/epjc/s10052-020-08713-1.
  33. J. Kvita, Study of methods of resolved top quark reconstruction in semileptonic t⁢t¯𝑡normal-¯𝑡t\bar{t}italic_t over¯ start_ARG italic_t end_ARG decay, Nucl. Instrum. Methods Phys. Res. A 900, 84 (2018), 10.1016/j.nima.2022.167172.
  34. ATLAS Collaboration, Evidence for the charge asymmetry in p⁢p→t⁢t¯normal-→𝑝𝑝𝑡normal-¯𝑡pp\rightarrow t\bar{t}italic_p italic_p → italic_t over¯ start_ARG italic_t end_ARG production at s=13𝑠13\sqrt{s}=13square-root start_ARG italic_s end_ARG = 13 TeV with the ATLAS detector (2022), 2208.12095.
  35. DØ Collaboration, Precise measurement of the top quark mass in dilepton decays using optimized neutrino weighting, Phys. Lett. B 752, 18 (2016), 10.1016/j.physletb.2015.10.086.
  36. CDF Collaboration, Measurement of the Top Quark Mass and t⁢t¯𝑡normal-¯𝑡t\overline{t}italic_t over¯ start_ARG italic_t end_ARG Production Cross Section from Dilepton Events at the Collider Detector at Fermilab, Phys. Rev. Lett. 80, 2779 (1998), 10.1103/PhysRevLett.80.2779.
  37. CMS Collaboration, Measurement of the top-quark mass in t⁢t¯𝑡normal-¯𝑡t\bar{t}italic_t over¯ start_ARG italic_t end_ARG events with dilepton final states in pp collisions at s=7⁢𝑇𝑒𝑉𝑠7𝑇𝑒𝑉\sqrt{s}=7\ \mbox{TeV}square-root start_ARG italic_s end_ARG = 7 TeV, Eur. Phys. J. C 72, 2202 (2012), 10.1140/epjc/s10052-012-2202-z.
  38. J. Alwall et al., The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations, JHEP 07, 79 (2014), 10.1007/jhep07(2014)079.
  39. P. Artoisenet et al., Automatic spin-entangled decays of heavy resonances in Monte Carlo simulations, JHEP 03, 15 (2013), 10.1007/jhep03(2013)015.
  40. T. Sjöstrand, S. Mrenna and P. Skands, A brief introduction to PYTHIA 8.1, Comput. Phys. Commun. 178, 852 (2008), 10.1016/j.cpc.2008.01.036.
  41. R. D. Ball et al., Parton distributions with LHC data, Nucl. Phys. B 867, 244 (2013), 10.1016/j.nuclphysb.2012.10.003.
  42. A. Buckley et al., LHAPDF6: parton density access in the LHC precision era, Eur. Phys. J. C 75, 132 (2015), 10.1140/epjc/s10052-015-3318-8.
  43. J. de Favereau et al., DELPHES 3, A modular framework for fast simulation of a generic collider experiment, JHEP 02, 057 (2014), 10.1007/jhep02(2014)057.
  44. The anti-kt jet clustering algorithm, JHEP 04, 063 (2008), 10.1088/1126-6708/2008/04/063.
  45. Fastjet user manual, Eur. Phys. J. C 72, 1 (2012), 10.1140/epjc/s10052-012-1896-2.
  46. M. Zaheer et al., Deep sets, In Proceedings of Advances in Neural Information Processing Systems, vol. 30. Long Beach, USA (2017), 1703.06114.
  47. D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, In Conference Track Proceedings of the 3rd International Conference on Learning Representations. San Diego, USA (2015), 1412.6980.
  48. A. Paszke et al., Pytorch: An imperative style, high-performance deep learning library, In Proceedings of Advances in Neural Information Processing Systems, vol. 32, pp. 8024–8035. Vancouver, Canada (2017), 1912.01703.
  49. C. Durkan et al., nflows: normalizing flows in PyTorch, 10.5281/zenodo.4296287 (2020).
  50. R. Girshick, Fast r-cnn, In Proceedings of The International Conference on Computer Vision, pp. 1440–1448. Santiago, Chile (2015), 1504.08083.
  51. CMS Collaboration, Measurement of Spin Correlations in t⁢t¯𝑡normal-¯𝑡t\bar{t}italic_t over¯ start_ARG italic_t end_ARG Production using the Matrix Element Method in the Muon+Jets Final State in p⁢p𝑝𝑝ppitalic_p italic_p Collisions at s=𝑠absent\sqrt{s}=square-root start_ARG italic_s end_ARG = 8 TeV, Phys. Lett. B 758, 321 (2016), 10.1016/j.physletb.2016.05.005.
  52. ATLAS Collaboration, Top-quark mass measurement in the all-hadronic t⁢t¯𝑡normal-¯𝑡t\bar{t}italic_t over¯ start_ARG italic_t end_ARG decay channel at s𝑠\sqrt{s}square-root start_ARG italic_s end_ARG = 8 TeV with the ATLAS detector, JHEP 09, 118 (2017), 10.1007/jhep09(2017)118.
  53. J. Erdmann et al., A likelihood-based reconstruction algorithm for top-quark pairs and the KLFitter framework, Nucl. Instrum. Meth. A 748, 18 (2014), 10.1016/j.nima.2014.02.029.
  54. ATLAS Collaboration, Search for the standard model Higgs boson produced in association with top quarks and decaying into a b⁢b¯𝑏normal-¯𝑏b\overline{b}italic_b over¯ start_ARG italic_b end_ARG pair in p⁢p𝑝𝑝ppitalic_p italic_p collisions at s=13⁢  ⁢TeV𝑠13  normal-TeV\sqrt{s}=13\text{ }\text{ }\mathrm{TeV}square-root start_ARG italic_s end_ARG = 13 roman_TeV with the ATLAS detector, Phys. Rev. D 97, 072016 (2018), 10.1103/PhysRevD.97.072016.
  55. J. Erdmann et al., From the bottom to the top - reconstruction of t⁢t¯𝑡normal-¯𝑡t{\bar{t}}italic_t over¯ start_ARG italic_t end_ARG events with deep learning, JINST 14, P11015 (2019), 10.1088/1748-0221/14/11/P11015.
  56. ATLAS Collaboration, C⁢P𝐶𝑃CPitalic_C italic_P Properties of Higgs Boson Interactions with Top Quarks in the t⁢t¯⁢H𝑡normal-¯𝑡𝐻t\overline{t}Hitalic_t over¯ start_ARG italic_t end_ARG italic_H and t⁢H𝑡𝐻tHitalic_t italic_H Processes Using H→γ⁢γnormal-→𝐻𝛾𝛾H\rightarrow\gamma\gammaitalic_H → italic_γ italic_γ with the ATLAS Detector, Phys. Rev. Lett. 125, 061802 (2020), 10.1103/PhysRevLett.125.061802.
  57. J. S. H. Lee et al., Zero-permutation jet-parton assignment using a self-attention network (2020), 2012.03542.
  58. A. Shmakov et al., Spanet: Generalized permutationless set assignment for particle physics using symmetry preserving attention, SciPost Phys. 12(5), 178 (2022), 10.21468/SciPostPhys.12.5.178.
  59. M. J. Fenton et al., Permutationless many-jet event reconstruction with symmetry preserving attention networks, Phys. Rev. D 105, 112008 (2022), 10.1103/PhysRevD.105.112008.
  60. ATLAS Collaboration, Measurements of top-quark pair single- and double-differential cross-sections in the all-hadronic channel in p⁢p𝑝𝑝ppitalic_p italic_p collisions at s=13𝑠13\sqrt{s}=13square-root start_ARG italic_s end_ARG = 13 TeV using the ATLAS detector, JHEP 01, 33 (2021), 10.1007/jhep01(2021)033.
  61. CDF Collaboration, Top quark mass measurement using the template method in the lepton+jetsnormal-leptonnormal-jets\mathrm{lepton}+\mathrm{jets}roman_lepton + roman_jets channel at CDF II, Phys. Rev. D 73, 032003 (2006), 10.1103/PhysRevD.73.032003.
  62. CMS Collaboration, Measurements of top-quark pair single- and double-differential cross-sections in the all-hadronic channel in p⁢p𝑝𝑝ppitalic_p italic_p collisions at s=13𝑠13\sqrt{s}=13square-root start_ARG italic_s end_ARG = 13 TeV using the ATLAS detector, Phys. Rev. D 93, 072004 (2016), 10.1103/PhysRevD.93.072004.
  63. A. L. Maas, A. Y. Hannun and A. Y. Ng, Rectifier nonlinearities improve neural network acoustic models, In Proceedings of The International Conference on Machine Learning, vol. 28, p. 3. PMLR, Atlanta, USA (2013).
  64. J. L. Ba, J. R. Kiros and G. E. Hinton, Layer normalization (2016), 1607.06450.
Citations (20)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now