Selecting interesting proton–proton collisions from the millions taking place each second in the Large Hadron Collider is a challenging task. A neural network optimized for a field-programmable gate array hardware enables 60 ns inference and reduces power consumption by a factor of 50.
This is a preview of subscription content, log in via an institution to check access.
References
Coelho, C. N. Jr et al. Nat. Mach. Intell. https://doi.org/10.1038/s42256-021-00356-5 (2021).
qkeras. GitHub https://github.com/google/qkeras (2021).
Duarte, J. et al. J. Instrum. 13, P07027 (2018).
Wang, K., Liu, Z., Lin, Y., Lin, J. & Han, S. Preprint at https://arxiv.org/abs/1811.08886 (2018).
TensorFlow Lite. TensorFlow https://www.tensorflow.org/lite (2021).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The author declares no competing interests.
Rights and permissions
About this article
Cite this article
Rousseau, D. Resource-efficient inference for particle physics. Nat Mach Intell 3, 656–657 (2021). https://doi.org/10.1038/s42256-021-00381-4
Published:
Issue Date:
DOI: https://doi.org/10.1038/s42256-021-00381-4
- Springer Nature Limited