“Scaling the Scattering Transform: Deep Hybrid Networks”, Edouard Oyallon, Eugene Belilovsky, Sergey Zagoruyko2017-03-27 (CNN)⁠:

[code] We use the scattering network as a generic and fixed initialization of the first layers of a supervised hybrid deep network. We show that early layers do not necessarily need to be learned, providing the best results to-date with pre-defined representations while being competitive with Deep CNNs.

Using a shallow cascade of 1×1 convolutions, which encodes scattering coefficients that correspond to spatial windows of very small sizes, permits to obtain AlexNet accuracy on the ImageNet ILSVRC2012_12ya. We demonstrate that this local encoding explicitly learns invariance with respect to rotations.

Combining scattering networks with a modern ResNet, we achieve a single-crop top 5 error of 11.4% on ImageNet ILSVRC2012_12ya, comparable to the Resnet-18 architecture, while using only 10 layers.

We also find that hybrid architectures can yield excellent performance in the small sample regime, exceeding their end-to-end counterparts, through their ability to incorporate geometrical priors.

We demonstrate this on subsets of the CIFAR-10 dataset and on the STL-10 dataset.