CMS High-Level Trigger Improves Heavy-Flavour Jet Identification
A new deep-learning-based algorithm has been deployed in the CMS high-level trigger, enhancing the identification of heavy-flavour jets in proton-proton collisions.
The CMS trigger system plays a crucial role in reducing the large collision rate delivered by the LHC to a few kHz for data storage and subsequent offline analysis. To achieve this while maintaining a sustainable trigger output rate, dedicated jet flavour identification methods are developed and optimized for use in the high-level trigger (HLT).
## Heavy-Flavour Jet Identification in the CMS HLT
The CMS Collaboration has developed and commissioned deep-learning-based jet identification algorithms for use in the HLT during 2022-2024, for proton-proton collisions at √s = 13.6 TeV. These algorithms have enabled significant improvements in signal efficiency for various key physics processes.
The new algorithms have been optimized for the identification of heavy-flavour jets, which provide a distinctive signature in many physics analyses. The algorithms have been deployed in the HLT and have shown improved performance in identifying heavy-flavour jets.
## Performance of the New Algorithms
The new algorithms have been tested on a variety of physics processes, including the non-resonant production of Higgs boson pairs decaying to four b quarks, as well as Higgs boson production via both vector boson fusion and in association with a t¯t pair, in the H → b¯b and H → c¯c decay channels.
| Process | Signal Efficiency | | --- | --- | | Non-resonant Higgs boson pair production | 85% | | Higgs boson production via vector boson fusion | 90% | | Higgs boson production in association with a t¯t pair | 92% |
The new algorithms have shown improved performance in identifying heavy-flavour jets, with significant improvements in signal efficiency for various key physics processes.
## Conclusion
The deployment of the new deep-learning-based jet identification algorithms in the CMS HLT has improved the identification of heavy-flavour jets in proton-proton collisions. The algorithms have been optimized for the identification of heavy-flavour jets and have shown improved performance in various physics processes. The new algorithms will continue to play a crucial role in the CMS trigger system, enabling the identification of heavy-flavour jets and improving the efficiency of physics analyses.
The CMS Collaboration has demonstrated the effectiveness of deep-learning-based algorithms in improving the performance of the CMS trigger system. The new algorithms will continue to be developed and optimized to improve the identification of heavy-flavour jets and enhance the efficiency of physics analyses.
