Zobrazeno 1 - 8
of 8
pro vyhledávání: '"Zarka, John"'
Publikováno v:
International Conference on Learning Representations, 2022
Deep convolutional classifiers linearly separate image classes and improve accuracy as depth increases. They progressively reduce the spatial dimension whereas the number of channels grows with depth. Spatial variability is therefore transformed into
Externí odkaz:
http://arxiv.org/abs/2110.05283
Numerical experiments demonstrate that deep neural network classifiers progressively separate class distributions around their mean, achieving linear separability on the training set, and increasing the Fisher discriminant ratio. We explain this mech
Externí odkaz:
http://arxiv.org/abs/2012.10424
We introduce a sparse scattering deep convolutional neural network, which provides a simple model to analyze properties of deep representation learning for classification. Learning a single dictionary matrix with a classifier yields a higher classifi
Externí odkaz:
http://arxiv.org/abs/1910.03561
Autor:
Andreux, Mathieu, Angles, Tomás, Exarchakis, Georgios, Leonarduzzi, Roberto, Rochette, Gaspar, Thiry, Louis, Zarka, John, Mallat, Stéphane, andén, Joakim, Belilovsky, Eugene, Bruna, Joan, Lostanlen, Vincent, Chaudhary, Muawiz, Hirn, Matthew J., Oyallon, Edouard, Zhang, Sixin, Cella, Carmine, Eickenberg, Michael
The wavelet scattering transform is an invariant 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 scatt
Externí odkaz:
http://arxiv.org/abs/1812.11214
Publikováno v:
International Conference on Learning Representations, 2022
International Conference on Learning Representations, 2022, Apr 2022, Online, France
International Conference on Learning Representations, 2022, Apr 2022, Online, France
Deep convolutional classifiers linearly separate image classes and improve accuracy as depth increases. They progressively reduce the spatial dimension whereas the number of channels grows with depth. Spatial variability is therefore transformed into
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::682f7390e49ce14cd30131d4273e876a
https://hal.science/hal-03373703
https://hal.science/hal-03373703
Publikováno v:
ICLR 2021-9th International Conference on Learning Representations
ICLR 2021-9th International Conference on Learning Representations, May 2021, Vienna / Virtual, Austria
ICLR 2021-9th International Conference on Learning Representations, May 2021, Vienna / Virtual, Austria
International audience; Numerical experiments demonstrate that deep neural network classifiers progressively separate class distributions around their mean, achieving linear separability on the training set, and increasing the Fisher discriminant rat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::36c63ea54c8a6f087dbe8970e1bee2e5
https://hal.archives-ouvertes.fr/hal-03169904
https://hal.archives-ouvertes.fr/hal-03169904
Publikováno v:
ICLR 2020-8th International Conference on Learning Representations
ICLR 2020-8th International Conference on Learning Representations, Apr 2020, Addis Ababa / Virtual, Ethiopia
ICLR 2020-8th International Conference on Learning Representations, Apr 2020, Addis Ababa / Virtual, Ethiopia
International audience; We introduce a sparse scattering deep convolutional neural network, which provides a simple model to analyze properties of deep representation learning for classification. Learning a single dictionary matrix with a classifier
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cd8fafa373bcc9cac1d341c12b61bfd0
https://inria.hal.science/hal-02976813
https://inria.hal.science/hal-02976813
Autor:
Andreux, Mathieu, Angles, Tomás, Exarchakis, Georgios, Leonarduzzi, Roberto, Rochette, Gaspar, Thiry, Louis, Zarka, John, Mallat, Stéphane, andén, Joakim, Belilovsky, Eugene, Bruna, Joan, Lostanlen, Vincent, Chaudhary, Muawiz, Hirn, Matthew J., Oyallon, Edouard, Zhang, Sixin, Cella, Carmine, Eickenberg, Michael
Publikováno v:
Journal of Machine Learning Research
Journal of Machine Learning Research, Microtome Publishing, 2020, 21 (60), pp.1-6
Journal of Machine Learning Research, 2020, 21 (60), pp.1-6
Journal of Machine Learning Research, Microtome Publishing, 2020, 21 (60), pp.1-6
Journal of Machine Learning Research, 2020, 21 (60), pp.1-6
International audience; The wavelet scattering transform is an invariant signal representation suitable for many signal processing and machine learning applications. We present the Kymatio software package, an easy-to-use, high-performance Python imp
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e89644ee2643c40a5fff944df8047747
https://hal.archives-ouvertes.fr/hal-02945354
https://hal.archives-ouvertes.fr/hal-02945354