Detecting an axion-like particle with machine learning at the LHC
Problem
Light axion-like particles (ALP mass MeV-GeV) are hard to detect because cosmological and astrophysical bounds on them are weak.
Why it matters
Axion-like particles are candidate new physics beyond the Standard Model, and detecting them would reshape fundamental physics.
What the AI did
A convolutional neural network tagging algorithm was trained on jet images (calorimeter towers as pixels) to identify highly boosted ALPs decaying into collimated photon pairs.
Results
Projected at 14 TeV HL-LHC with 3000 fb^-1: sensitivity extended to ALP masses 0.3-5 GeV, with bounds stronger than existing ALP-photon coupling limits.
Impact
Projected sensitivity gains could strengthen future ALP searches at the HL-LHC.
Record
| Venue | Journal of High Energy Physics |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Jie Ren, Daohan Wang, Lei Wu, Jin Min Yang, Mengchao Zhang |
| Citations | 31 (OpenAlex, at import time) |
| References | 87 |
| Model or method | convolutional neural network (CNN) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1007/jhep11(2021)138 |
| Open access | diamond |
| Keywords | axion-like particles, CNN, jet images, LHC, machine learning, tagging algorithm |
Importance rating
significant — thousands to millions affected
ALPs are candidate new physics beyond the Standard Model; extending HL-LHC sensitivity to an unstudied mass range serves particle physics, a foundational scientific field, though results are projected.
More in Physics & Astronomy
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Magnetic control of tokamak plasmas through deep reinforcement learning | 2022 | 739 |
7
|
| 2 | Solving the quantum many-body problem with artificial neural networks | 2017 | 2,205 |
7
|
| 3 | Learning high-accuracy error decoding for quantum processors | 2024 | 90 |
7
|
| 4 | Optimizing Quantum Error Correction Codes with Reinforcement Learning | 2019 | 158 |
7
|
| 5 | Machine-learning-assisted correction of correlated qubit errors in a topological code | 2018 | 141 |
7
|