Spectral representations for convolutional neural networks
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
Spectral representations for convolutional neural networks
O Rippel, J Snoek, RP Adams - … of the 28th International Conference on …, 2015 - dl.acm.org
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - arXiv preprint arXiv:1506.03767, 2015 - arxiv.org
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
[PDF][PDF] Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - 2015 - people.ee.duke.edu
They are computationally expensive Any pooling step reduces the dimensionality by at least
4 Previous work suggests using FFTs to compute the convolutional mask–even for small …
4 Previous work suggests using FFTs to compute the convolutional mask–even for small …
[PDF][PDF] Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - stat, 2015 - cs.princeton.edu
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
Spectral representations for convolutional neural networks
O Rippel, J Snoek, RP Adams - Advances in Neural …, 2015 - collaborate.princeton.edu
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
[PDF][PDF] Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - stat, 2015 - lips.cs.princeton.edu
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
[PDF][PDF] Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - proceedings.neurips.cc
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - arXiv e-prints, 2015 - ui.adsabs.harvard.edu
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
[PDF][PDF] Spectral Representations for Convolutional Neural Networks
O Rippel, J Snoek, RP Adams - stat, 2015 - cs.princeton.edu
Discrete Fourier transforms provide a significant speedup in the computation of convolutions
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …
in deep learning. In this work, we demonstrate that, beyond its advantages for efficient …