• Voir en Machine learning to reveal more about LHC particle collisions The CMS Collaboration demonstrates that machine learning can outperform traditional methods in the full reconstruction of particle collisions at the LHC 18 February, 2026 | ByCMS collaboration A particle collision reconstructed using the new CMS machine-learning-based particle-flow (MLPF) algorithm. • The HFEM and HFHAD signals come from the forward calorimeters, which measure energy from particles travelling close to the beamline. • (Image: CMS) TheCMSCollaboration has shown, for the first time, that machine learning can be used to fully reconstruct particle collisions at theLHC. • Thisnew approachcan reconstruct collisions more quickly and precisely than traditional methods, helping physicists better understand LHC data. • Each proton-proton collision at the LHC sprays out a complex pattern of particles that must be carefully reconstructed to allow physicists to study what really happened. • For more than a decade, CMS has used a particle-flow (PF) algorithm, which combines information from the experiment’s differentdetectors, to identify each particle produced in a collision.
Article Summaries:
- CMS has demonstrated that a machine‑learning algorithm can fully reconstruct proton‑proton collisions at the Large Hadron Collider (LHC) with speed and precision comparable to, or better than, the long‑used particle‑flow (PF) method. The new machine‑learning‑based particle‑flow (MLPF) model replaces hand‑crafted reconstruction rules with a single neural network trained on simulated events, learning directly how particles appear in the detectors. Benchmarks show MLPF matches traditional performance and improves jet‑momentum resolution by 10-20 % for top‑quark events. Running on GPUs, it reconstructs collisions faster than CPU‑based PF, promising greater efficiency for the upcoming High‑Luminosity LHC era.
Sources: