Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2306.04712 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910282849189888 |
|---|---|
| author | Shenoy, Rohan Duarte, Javier Herwig, Christian Hirschauer, James Noonan, Daniel Pierini, Maurizio Tran, Nhan Suarez, Cristina Mantilla |
| author_facet | Shenoy, Rohan Duarte, Javier Herwig, Christian Hirschauer, James Noonan, Daniel Pierini, Maurizio Tran, Nhan Suarez, Cristina Mantilla |
| contents | 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) to learn a differentiable, fast approximation of the EMD and demonstrate that it can be used as a substitute for computing-intensive EMD implementations. We apply this differentiable approximation in the training of an autoencoder-inspired neural network (encoder NN) for data compression at the high-luminosity LHC at CERN. The goal of this encoder NN is to compress the data while preserving the information related to the distribution of energy deposits in particle detectors. We demonstrate that the performance of our encoder NN trained using the differentiable EMD CNN surpasses that of training with loss functions based on mean squared error. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_04712 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Differentiable Earth Mover's Distance for Data Compression at the High-Luminosity LHC Shenoy, Rohan Duarte, Javier Herwig, Christian Hirschauer, James Noonan, Daniel Pierini, Maurizio Tran, Nhan Suarez, Cristina Mantilla High Energy Physics - Experiment Machine Learning Instrumentation and Detectors 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) to learn a differentiable, fast approximation of the EMD and demonstrate that it can be used as a substitute for computing-intensive EMD implementations. We apply this differentiable approximation in the training of an autoencoder-inspired neural network (encoder NN) for data compression at the high-luminosity LHC at CERN. The goal of this encoder NN is to compress the data while preserving the information related to the distribution of energy deposits in particle detectors. We demonstrate that the performance of our encoder NN trained using the differentiable EMD CNN surpasses that of training with loss functions based on mean squared error. |
| title | Differentiable Earth Mover's Distance for Data Compression at the High-Luminosity LHC |
| topic | High Energy Physics - Experiment Machine Learning Instrumentation and Detectors |
| url | https://arxiv.org/abs/2306.04712 |