Stephen P. W. Jiggins

Gebäude 1, 2.0G, 2.307

Notkestraße 85

Hamburg, 22607, Germany

I am a machine learning researcher and software developer with 10+ years bridging ML research and scientific applications across particle physics, nuclear, medical, and automotive domains. I am a Research Fellow at DESY / ATLAS, Principal Investigator of the Helmholtz AI project NEEDLE, and Lead AI Engineer at useblocks GmbH.

Current work

Leading NEEDLE — diffusion/flow-matching neural simulation-based inference for distributed HPC (Helmholtz AI, €0.5M grant): NEEDLE SBI
Generative Machine Learning for simulating pile-up (noise) at the Large Hadron Collider (LHC) at CERN using diffusion and flow matching models
Probabilistic Machine Learning for Monte Carlo physics simulations and neural simulation-based inferencing
Agentic AI and LLM tooling for automotive software-requirements traceability (useblocks GmbH).

Background Previously a postdoctoral researcher at the University of Edinburgh (2021–2022) and Universität Freiburg (2018–2022), after a PhD in High Energy Physics at University College London / CERN (2013–2017), and BSc/MSci degrees at King’s College London.

news

May 28, 2026	First look at NEEDLE-SBI framework at 28th Conference on Computing in High Energy and Nuclear Physics (CHEP 2026) by K.Schmidt.
May 26, 2026	Alpha version of NEEDLE-SBI released on github.
Oct 28, 2025	Paper accepted: Neural Quasiprobabilistic Likelihood Ratio Estimation with Negatively Weighted Data is now live on the IOP Machine Learning: Science Technology journal.
Aug 04, 2025	Started as Lead AI Engineer at useblocks GmbH (Munich), working on agentic AI for automotive software-requirements traceability.
Oct 14, 2024	New paper: Neural Quasiprobabilistic Likelihood Ratio Estimation with Negatively Weighted Data - Extending likelihood ratio estimation using neural surrogates to quasi-probabilistic densities (negative probablities).

selected publications

MLST

Neural quasiprobabilistic likelihood ratio estimation with negatively weighted data

Matthew Drnevich, Stephen Jiggins, Judith Katzy, and 1 more author

Machine Learning: Science and Technology, Oct 2025

DOI Bib

@article{Drnevich_2025,
  doi = {10.1088/2632-2153/ae0def},
  url = {https://doi.org/10.1088/2632-2153/ae0def},
  year = {2025},
  month = oct,
  publisher = {IOP Publishing},
  volume = {6},
  number = {4},
  pages = {045023},
  author = {Drnevich, Matthew and Jiggins, Stephen and Katzy, Judith and Cranmer, Kyle},
  title = {Neural quasiprobabilistic likelihood ratio estimation with negatively weighted data},
  journal = {Machine Learning: Science and Technology},
}

arXiv

Quasiprobabilistic Density Ratio Estimation with a Reverse Engineered Classification Loss Function

Matthew Drnevich, Stephen Jiggins, and Kyle Cranmer

Oct 2025

arXiv Bib

@misc{drnevich2025quasiprobabilisticdensityratioestimation,
  title = {Quasiprobabilistic Density Ratio Estimation with a Reverse Engineered Classification Loss Function},
  author = {Drnevich, Matthew and Jiggins, Stephen and Cranmer, Kyle},
  year = {2025},
  eprint = {2512.19913},
  archiveprefix = {arXiv},
  primaryclass = {stat.ML},
  url = {https://arxiv.org/abs/2512.19913},
}

EPJP

Quantum information meets high-energy physics: input to the update of the European strategy for particle physics

Yoav Afik, Federica Fabbri, Matthew Low, and 68 more authors

The European Physical Journal Plus, Sep 2025

DOI Bib

@article{Afik2025,
  author = {Afik, Yoav and Fabbri, Federica and Low, Matthew and Marzola, Luca and Aguilar-Saavedra, Juan Antonio and Altakach, Mohammad Mahdi and Asbah, Nedaa Alexandra and Bai, Yang and Banks, Hannah and Barr, Alan J. and Bernal, Alexander and Browder, Thomas E. and Caban, Pawe{\l} and Casas, J. Alberto and Cheng, Kun and D{\'e}liot, Fr{\'e}d{\'e}ric and Demina, Regina and Di Domenico, Antonio and Eckstein, Micha{\l} and Fabbrichesi, Marco and Fuks, Benjamin and Gabrielli, Emidio and Gon{\c{c}}alves, Dorival and Grabarczyk, Rados{\l}aw and Grossi, Michele and Han, Tao and Hobbs, Timothy J. and Horodecki, Pawe{\l} and Howarth, James and Hsu, Shih-Chieh and Jiggins, Stephen and Jones, Eleanor and Jung, Andreas W. and Knue, Andrea Helen and Korn, Steffen and Lagouri, Theodota and Lamba, Priyanka and Landi, Gabriel T. and Li, Haifeng and Li, Qiang and Low, Ian and Maltoni, Fabio and McFayden, Josh and McGinnis, Navin and Morales, Roberto A. and Moreno, Jes{\'u}s M. and de Nova, Juan Ramonoz and Negro, Giulia and Pagani, Davide and Pelliccioli, Giovanni and Pinamonti, Michele and Pintucci, Laura and Ravina, Baptiste and Ruzi, Alim and Sakurai, Kazuki and Simpson, Ethan and Sioli, Maximiliano and Su, Shufang and Trifinopoulos, Sokratis and Vahsen, Sven E. and Vallecorsa, Sofia and Vicini, Alessandro and Vos, Marcel and Vryonidou, Eleni and White, Chris D. and White, Martin J. and Wildridge, Andrew J. and Wu, Tong Arthur and Zani, Laura and Zhang, Yulei and Zoch, Knut},
  title = {Quantum information meets high-energy physics: input to the update of the European strategy for particle physics},
  journal = {The European Physical Journal Plus},
  year = {2025},
  month = sep,
  day = {09},
  volume = {140},
  number = {9},
  pages = {855},
  issn = {2190-5444},
  doi = {10.1140/epjp/s13360-025-06752-9},
  url = {https://doi.org/10.1140/epjp/s13360-025-06752-9},
}

JHEP
Measurements of WH and ZH production with Higgs boson decays into bottom quarks and direct constraints on the charm Yukawa coupling in 13 TeV pp collisions with the ATLAS detector

ATLAS Collaboration

Journal of High Energy Physics, Apr 2025

Abs DOI Bib

A study of the Higgs boson decaying into bottom quarks (H \textrightarrow }} b\backslashoverline{b} }}) and charm quarks (H \textrightarrow }} c\backslashoverline{c} }}) is performed, in the associated production channel of the Higgs boson with a W or Z boson, using 140 fb−1 of proton-proton collision data at }} \backslashsqrt{s} }}= 13 TeV collected by the ATLAS detector. The individual production of WH and ZH with H \textrightarrow }} b\backslashoverline{b} }}is established with observed (expected) significances of 5.3 (5.5) and 4.9 (5.6) standard deviations, respectively. Differential cross-section measurements of the gauge boson transverse momentum within the simplified template cross-section framework are performed in a total of 13 kinematical fiducial regions.
@article{Aad2025_WHZH_2025, author = {Collaboration, ATLAS}, title = {Measurements of WH and ZH production with Higgs boson decays into bottom quarks and direct constraints on the charm Yukawa coupling in 13 TeV pp collisions with the ATLAS detector}, journal = {Journal of High Energy Physics}, year = {2025}, month = apr, day = {10}, volume = {2025}, number = {4}, pages = {75}, issn = {1029-8479}, doi = {10.1007/JHEP04(2025)075}, url = {https://doi.org/10.1007/JHEP04(2025)075}, }

PUB

Monte Carlo modelling of loop-induced ZH production in ATLAS

Apr 2022

All figures including auxiliary figures are available at https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2022-055

Bib HTML

@techreport{ATL-PHYS-PUB-2022-055,
  collaboration = {ATLAS},
  title = {{Monte Carlo modelling of loop-induced $ZH$ production in
                         ATLAS}},
  institution = {CERN},
  reportnumber = {ATL-PHYS-PUB-2022-055},
  address = {Geneva},
  year = {2022},
  url = {https://cds.cern.ch/record/2842239},
  note = {All figures including auxiliary figures are available at
                         https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2022-055},
}

PoS

Development of novel experimental techniques to improve the understanding of the Higgs sector by the ATLAS experiment

Stephen Jiggins

PoS, Apr 2022

DOI Bib HTML

@article{Jiggins:2022ZT,
  author = {Jiggins, Stephen},
  title = {{Development of novel experimental techniques to improve the understanding of the Higgs sector by the ATLAS experiment}},
  doi = {10.22323/1.414.0526},
  journal = {PoS},
  year = {2022},
  volume = {ICHEP2022},
  pages = {526},
}

JHEP
Modelling and computational improvements to the simulation of single vector-boson plus jet processes for the ATLAS experiment

ATLAS Collaboration

Journal of High Energy Physics, Aug 2022

Abs DOI Bib

This paper presents updated Monte Carlo configurations used to model the production of single electroweak vector bosons (W, Z/γ∗) in association with jets in proton-proton collisions for the ATLAS experiment at the Large Hadron Collider. Improvements pertaining to the electroweak input scheme, parton-shower splitting kernels and scale-setting scheme are shown for multi-jet merged configurations accurate to next-to-leading order in the strong and electroweak couplings. The computational resources required for these set-ups are assessed, and approximations are introduced resulting in a factor three reduction of the per-event CPU time without affecting the physics modelling performance. Continuous statistical enhancement techniques are introduced by ATLAS in order to populate low cross-section regions of phase space and are shown to match or exceed the generated effective luminosity. This, together with the lower per-event CPU time, results in a 50% reduction in the required computing resources compared to a legacy set-up previously used by the ATLAS collaboration. The set-ups described in this paper will be used for future ATLAS analyses and lay the foundation for the next generation of Monte Carlo predictions for single vector-boson plus jets production.
@article{Aad2022, author = {Collaboration, ATLAS}, title = {Modelling and computational improvements to the simulation of single vector-boson plus jet processes for the ATLAS experiment}, journal = {Journal of High Energy Physics}, year = {2022}, month = aug, day = {05}, volume = {2022}, number = {8}, pages = {89}, issn = {1029-8479}, doi = {10.1007/JHEP08(2022)089}, url = {https://doi.org/10.1007/JHEP08(2022)089}, }