IMEXnet: A Forward Stable Deep Neural Network

TL;DR

IMEXnet is a novel deep neural network architecture inspired by semi-implicit PDE methods, enhancing stability, robustness, and field of view in image processing tasks, demonstrated on semantic segmentation datasets.

Contribution

Introduces IMEXnet, a stable and efficient neural network leveraging semi-implicit methods to improve robustness and address field of view limitations in convolutional networks.

Findings

IMEXnet shows increased stability over residual networks.

The implicit step enhances the field of view in the network.

Effective performance demonstrated on NYU Depth dataset.

Abstract

Deep convolutional neural networks have revolutionized many machine learning and computer vision tasks, however, some remaining key challenges limit their wider use. These challenges include improving the network's robustness to perturbations of the input image and the limited ``field of view'' of convolution operators. We introduce the IMEXnet that addresses these challenges by adapting semi-implicit methods for partial differential equations. Compared to similar explicit networks, such as residual networks, our network is more stable, which has recently shown to reduce the sensitivity to small changes in the input features and improve generalization. The addition of an implicit step connects all pixels in each channel of the image and therefore addresses the field of view problem while still being comparable to standard convolutions in terms of the number of parameters and computational complexity. We also present a new dataset for semantic segmentation and demonstrate the effectiveness of our architecture using the NYU Depth dataset.