Speaker
Denis-Patrick Odagiu
(ETH Zurich (CH))
Description
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 LHC during its high-luminosity phase.
Through quantization-aware training and efficient synthetization for a specific field programmable gate array, we show that $\mathcal{O}(100)$ ns inference of complex architectures such as Deep Sets and Interaction Networks is feasible at a relatively low computational resource cost.
| What of the following keywords match your abstract best? | FPGAs |
|---|---|
| Please tick if you are a PhD student and wish to take part to the poster prize competition! | I am a PhD student |
Author
Denis-Patrick Odagiu
(ETH Zurich (CH))
Co-authors
Andre Sznajder
(Universidade do Estado do Rio de Janeiro (BR))
Javier Mauricio Duarte
(Univ. of California San Diego (US))
Thea Aarrestad
(ETH Zurich (CH))
Zhiqiang (Walkie) Que
(Imperial College London)
