TY - JOUR
T1 - Kymatio
T2 - Scattering transforms in python
AU - Andreux, Mathieu
AU - Angles, Tomás
AU - Exarchakis, Georgios
AU - Leonarduzzi, Roberto
AU - Rochette, Gaspar
AU - Thiry, Louis
AU - Zarka, John
AU - Mallat, Stéphane
AU - Andén, Joakim
AU - Belilovsky, Eugene
AU - Bruna, Joan
AU - Lostanlen, Vincent
AU - Chaudhary, Muawiz
AU - Hirn, Matthew J.
AU - Oyallon, Edouard
AU - Zhang, Sixin
AU - Cella, Carmine
AU - Eickenberg, Michael
N1 - Funding Information:
We thank Laurent Sifre, Sergey Zagoruyko and Gabriel Huang for their helpful comments. The project was supported by ERC InvariantClass 320959. EB is funded by a Google Focused Research Award and IVADO. MJH is partially supported by Alfred P. Sloan Fellowship #FG-2016-6607, DARPA Young Faculty Award #D16AP00117, and NSF grant #1620216. The Flatiron Institute is a division of the Simons Foundation.
Publisher Copyright:
©2020 Mathieu Andreux, Tomás Angles, Georgios Exarchakis, Roberto Leonarduzzi, Gaspar Rochette, Louis Thiry, John Zarka, Stéphane Mallat, Joakim Andén, Eugene Belilovsky, Joan Bruna, Vincent Lostanlen, Muawiz Chaudhary, Matthew J. Hirn, Edouard Oyallon, Sixin Zhang, Carmine Cella, Michael Eickenberg.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - The wavelet scattering transform is an invariant and stable signal representation suitable for many signal processing and machine learning applications. We present the Kymatio software package, an easy-to-use, high-performance Python implementation of the scattering transform in 1D, 2D, and 3D that is compatible with modern deep learning frameworks, including PyTorch and TensorFlow/Keras. The transforms are implemented on both CPUs and GPUs, the latter offering a significant speedup over the former. The package also has a small memory footprint. Source code, documentation, and examples are available under a BSD license at https://www.kymat.io.
AB - The wavelet scattering transform is an invariant and stable signal representation suitable for many signal processing and machine learning applications. We present the Kymatio software package, an easy-to-use, high-performance Python implementation of the scattering transform in 1D, 2D, and 3D that is compatible with modern deep learning frameworks, including PyTorch and TensorFlow/Keras. The transforms are implemented on both CPUs and GPUs, the latter offering a significant speedup over the former. The package also has a small memory footprint. Source code, documentation, and examples are available under a BSD license at https://www.kymat.io.
KW - Convolutional Networks
KW - GPUs
KW - Invariance
KW - Scattering Transform
KW - Wavelets
UR - http://www.scopus.com/inward/record.url?scp=85087199310&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85087199310&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:85087199310
SN - 1532-4435
VL - 21
JO - Journal of Machine Learning Research
JF - Journal of Machine Learning Research
ER -