DeepAI AI Chat
Log In Sign Up

Generalized Dilation Neural Networks

by   Gavneet Singh Chadha, et al.
Fachhochschule Südwestfalen

Vanilla convolutional neural networks are known to provide superior performance not only in image recognition tasks but also in natural language processing and time series analysis. One of the strengths of convolutional layers is the ability to learn features about spatial relations in the input domain using various parameterized convolutional kernels. However, in time series analysis learning such spatial relations is not necessarily required nor effective. In such cases, kernels which model temporal dependencies or kernels with broader spatial resolutions are recommended for more efficient training as proposed by dilation kernels. However, the dilation has to be fixed a priori which limits the flexibility of the kernels. We propose generalized dilation networks which generalize the initial dilations in two aspects. First we derive an end-to-end learnable architecture for dilation layers where also the dilation rate can be learned. Second we break up the strict dilation structure, in that we develop kernels operating independently in the input space.


page 1

page 2

page 3

page 4


Anomaly Detection And Classification In Time Series With Kervolutional Neural Networks

Recently, with the development of deep learning, end-to-end neural netwo...

Irregular Convolutional Neural Networks

Convolutional kernels are basic and vital components of deep Convolution...

Convolutional Neural Networks for Time-dependent Classification of Variable-length Time Series

Time series data are often obtained only within a limited time range due...

Convolutional Shapelet Transform: A new approach for time series shapelets

Shapelet-based algorithms are widely used for time series classification...

Channel-Based Attention for LCC Using Sentinel-2 Time Series

Deep Neural Networks (DNNs) are getting increasing attention to deal wit...

What Can ResNet Learn Efficiently, Going Beyond Kernels?

How can neural networks such as ResNet efficiently learn CIFAR-10 with t...

On the Spectral Bias of Convolutional Neural Tangent and Gaussian Process Kernels

We study the properties of various over-parametrized convolutional neura...