Machine Learning for Invisible Dark Boson Searches at the Electron-Ion Collider

Abstract

We investigate whether adding t, the positive magnitude of the squared nuclear four-momentum transfer, enables boosted decision trees (BDTs) to improve invisible-dark-boson selection relative to optimized rectangular cuts at the Electron-Ion Collider. We model coherent exclusive scalar and vector production at generator level in electron-gold collisions at 18 GeV by 100 GeV per nucleon. Both methods use identical weighted samples, inputs, preselection, and optimization objectives. Using only electron information, the BDT provided no consistent advantage over optimized cuts for signal selection across 11 masses for each boson type. When both methods also use t, the BDT distinguishes signal from background slightly better than optimized cuts at 10 GeV for both boson types. These results motivate further investigation of machine learning in EIC dark-boson searches through exclusive processes where t can be reconstructed.

Full-event anomaly detection for new physics searches at the Electron-Ion Collider
Dimitrios Athanasakos, Sebastian Grieninger, Hongkai Liu +3
arXiv8 Oct 2026PreprintarXiv licenceFull textComputational physics High-energy & particle physicsarXiv
Fast prediction of electron-beam parameters using transfer maps and machine learning in a digital model of a MeV γγ collider
Jiawei Zhang, Yongsheng Huang, Yuancun Nie +3
arXiv5 Nov 2026PreprintarXiv licenceFull textApplied physics & instrumentationarXiv
Invisible dark-sector radiation: analytic resummation for dark photon searches
Xiangyu Deng, Rongfeng Ou, Yi Li +1
arXiv12 Oct 2026PreprintarXiv licenceFull textHigh-energy & particle physicsarXiv
Fast BIB simulation at a future Muon Collider with generative machine learning
Radha Mastandrea, Shiyu Peng, Benjamin Rosser +1
arXiv10 Sep 2026PreprintCC BY 4.0Full textApplied physics & instrumentation High-energy & particle physicsarXiv
Probing lepton flavor mixing in W_R searches with machine learning at the LHC
Jin-Man Cai, Gang Li
arXiv21 Sep 2026PreprintarXiv licenceFull textHigh-energy & particle physicsarXiv

Cite

BibTeX RIS

BibTeX

@misc{mighty2026machine,
  title = {{Machine Learning for Invisible Dark Boson Searches at the Electron-Ion Collider}},
  author = {Rojae Mighty and Ankush Reddy Kanuganti},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.22484},
  eprint = {2609.22484},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.22484},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Machine Learning for Invisible Dark Boson Searches at the Electron-Ion Collider
AU  - Rojae Mighty
AU  - Ankush Reddy Kanuganti
PY  - 2026
DA  - 2026/09/18
PB  - arXiv
DO  - 10.48550/arxiv.2609.22484
UR  - https://arxiv.org/abs/2609.22484
AB  - We investigate whether adding t, the positive magnitude of the squared nuclear four-momentum transfer, enables boosted decision trees (BDTs) to improve invisible-dark-boson selection relative to optimized rectangular cuts at the Electron-Ion Collider. We model coherent exclusive scalar and vector production at generator level in electron-gold collisions at 18 GeV by 100 GeV per nucleon. Both methods use identical weighted samples, inputs, preselection, and optimization objectives. Using only electron information, the BDT provided no consistent advantage over optimized cuts for signal selection across 11 masses for each boson type. When both methods also use t, the BDT distinguishes signal from background slightly better than optimized cuts at 10 GeV for both boson types. These results motivate further investigation of machine learning in EIC dark-boson searches through exclusive processes where t can be reconstructed.
ER  -

Abstract

Related

Cite