Ultrafast jet classification at the HL-LHC
Open Access
- 18 July 2024
- journal article
- research article
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 5 (3) , 035017
Abstract
Three machine learning models are used to perform jet origin classification. These models are optimized for deployment on a field-programmable gate array device. In this context, we demonstrate how latency and resource consumption scale with the input size and choice of algorithm. Moreover, the models proposed here are designed to work on the type of data and under the foreseen conditions at the CERN large hadron collider during its high-luminosity phase. Through quantization-aware training and efficient synthetization for a specific field programmable gate array, we show that O ( 100 ) ns inference of complex architectures such as Deep Sets and Interaction Networks is feasible at a relatively low computational resource cost.Keywords
All Related Versions
Funding Information
- NSF
- High Energy Physics (DE-SC0021187)
- U.S. Department of Energy
- European Union (772369)
- European Research Council
- EPSRC (EP/L016796/1)
- High Energy Physics “Designing efficient edge AI with physics phenomena” (DE-FOA-0002705)
- Department of Energy
- Swiss National Science Foundation (201594)
- Hamburg University
- CAPES
- A3D3 Institute (OAC-2117997)
- Fermi Research Alliance, LLC (DE-AC02-07CH11359)
- Germany’s Excellence Strategy (390833306)
- DFG, German Research Foundation
- Deutsche Forschungsgemeinschaft
This publication has 25 references indexed in Scilit:
- Nanosecond machine learning regression with deep boosted decision trees in FPGA for high energy physicsJournal of Instrumentation, 2022
- An efficient Lorentz equivariant graph neural network for jet taggingJournal of High Energy Physics, 2022
- Nanosecond machine learning event classification with boosted decision trees in FPGA for high energy physicsJournal of Instrumentation, 2021
- Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectorsNature Machine Intelligence, 2021
- Fast inference of Boosted Decision Trees in FPGAs for particle physicsJournal of Instrumentation, 2020
- Graph neural networks: A review of methods and applicationsAI Open, 2020
- Energy flow networks: deep sets for particle jetsJournal of High Energy Physics, 2019
- Fast inference of deep neural networks in FPGAs for particle physicsJournal of Instrumentation, 2018
- The importance of calorimetry for highly-boosted jet substructureJournal of Instrumentation, 2018
- The anti-ktjet clustering algorithmJournal of High Energy Physics, 2008