Differentiable Earth mover’s distance for data compression at the high-luminosity LHC

Differentiable Earth mover’s distance for data compression at the high-luminosity LHC

The Earth mover’s distance (EMD) is a useful metric for image recognition and classification, but its usual implementations are not differentiable or too slow to be used as a loss function for training other algorithms via gradient descent. In this paper, we train a convolutional neural network (CNN...

Volledige beschrijving

Bibliografische gegevens
Hoofdauteurs: Rohan Shenoy, Javier Duarte, Christian Herwig, James Hirschauer, Daniel Noonan, Maurizio Pierini, Nhan Tran, Cristina Mantilla Suarez
Formaat: Artikel
Taal:English
Gepubliceerd in: IOP Publishing 2023-01-01
Reeks:Machine Learning: Science and Technology
Onderwerpen:
Earth movers distance
data compression
convolutional neural networks
high energy physics
Online toegang:https://doi.org/10.1088/2632-2153/ad1139

Internet

https://doi.org/10.1088/2632-2153/ad1139

Gelijkaardige items