Speaker
Description
The Large Hadron Collider (LHC) collides protons at a rate of 40 million collisions per second. To filter the massive amount of data for interesting physics, the real-time trigger systems inside detectors at the LHC necessitate smart and sophisticated triggers that are 1) efficient enough to simultaneously reject large backgrounds and keep enough signal, 2) compact enough to meet hardware constraints, and 3) fast enough to meet the latency requirements. We present an implementation of a particle jet tagger using the High-Granularity Quantization (HGQ) library, trained and validated on data simulating the conditions of the LHC. Subsequently, the model is converted into high level synthesis using the hls4ml hardware-software codesign tool to be implemented onto real-time trigger hardware. This work explores the differences between the QKeras and HGQ model implementations, including architecture differences, jet tagging efficiencies, and hardware capabilities.
| Do you plan to submit a 4-page extended abstract on OpenReview (only for Presentations/Posters)? | No |
|---|