Translating Diffusion, Wavelets, and Regularisation into Residual Networks
Alt, Tobias, Weickert, Joachim, Peter, Pascal
Convolutional neural networks (CNNs) often perform well, but their stability is poorly understood. To address this problem, we consider the simple prototypical problem of signal denoising, where classical approaches such as nonlinear diffusion, wavelet-based methods and regularisation offer provable stability guarantees. To transfer such guarantees to CNNs, we interpret numerical approximations of these classical methods as a specific residual network (ResNet) architecture. This leads to a dictionary which allows to translate diffusivities, shrinkage functions, and regularisers into activation functions, and enables a direct communication between the four research communities. On the CNN side, it does not only inspire new families of nonmonotone activation functions, but also introduces intrinsically stable architectures for an arbitrary number of layers.
Feb-7-2020
- Country:
- Oceania > Australia
- New South Wales > Sydney (0.04)
- North America
- United States
- New York (0.04)
- Texas > Travis County
- Austin (0.04)
- Nevada > Clark County
- Las Vegas (0.04)
- California > San Diego County
- San Diego (0.04)
- Colorado > Denver County
- Denver (0.04)
- Hawaii > Honolulu County
- Honolulu (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Utah > Salt Lake County
- Salt Lake City (0.04)
- Massachusetts > Middlesex County
- Cambridge (0.04)
- Canada
- Quebec > Montreal (0.04)
- British Columbia > Metro Vancouver Regional District
- Vancouver (0.04)
- United States
- Europe
- Sweden > Stockholm
- Stockholm (0.04)
- Germany
- Saarland > Saarbrücken (0.04)
- Baden-Württemberg > Stuttgart Region
- Stuttgart (0.04)
- Sweden > Stockholm
- Asia > Middle East
- Israel > Haifa District > Haifa (0.04)
- Oceania > Australia
- Genre:
- Research Report (0.64)
- Technology: