249 citations · 699 across the 39 of their papers we have counts for
8 papers · 1 filter
Greedy Layerwise Learning Can Scale to ImageNet
Eugene Belilovsky, Michael Eickenberg, Edouard Oyallon
Shallow supervised 1-hidden layer neural networks have a number of favorable properties that make them easier to interpret, analyze, and optimize than their deep counterparts, but…
Kymatio: Scattering Transforms in Python
Mathieu Andreux, Tomás Angles, Georgios Exarchakis +15
The wavelet scattering transform is an invariant signal representation suitable for many signal processing and machine learning applications. We present the Kymatio software packag…
On Lazy Training in Differentiable Programming
Lenaic Chizat, Edouard Oyallon, Francis Bach
In a series of recent theoretical works, it was shown that strongly over-parameterized neural networks trained with gradient-based methods could converge exponentially fast to zero…
Compressing the Input for CNNs with the First-Order Scattering Transform
Edouard Oyallon, Eugene Belilovsky, Sergey Zagoruyko +1
We study the first-order scattering transform as a candidate for reducing the signal processed by a convolutional neural network (CNN). We show theoretical and empirical evidence t…
Scattering Networks for Hybrid Representation Learning
Edouard Oyallon, Sergey Zagoruyko, Gabriel Huang +4
Scattering networks are a class of designed Convolutional Neural Networks (CNNs) with fixed weights. We argue they can serve as generic representations for modelling images. In par…
Nonlinear Acceleration of CNNs
Damien Scieur, Edouard Oyallon, Alexandre d'Aspremont +1
The Regularized Nonlinear Acceleration (RNA) algorithm is an acceleration method capable of improving the rate of convergence of many optimization schemes such as gradient descend,…
Online Regularized Nonlinear Acceleration
Damien Scieur, Edouard Oyallon, Alexandre d'Aspremont +1
Regularized nonlinear acceleration (RNA) estimates the minimum of a function by post-processing iterates from an algorithm such as the gradient method. It can be seen as a regulari…
i-RevNet: Deep Invertible Networks
Jörn-Henrik Jacobsen, Arnold Smeulders, Edouard Oyallon
It is widely believed that the success of deep convolutional networks is based on progressively discarding uninformative variability about the input with respect to the problem at…