The delta operator approach to continuous-time cellular neural networks (CT-CNNS) is investigated in terms of a robust realization. It is shown that earlier results concerning the robustness of CTCNNS can be obtained as a limiting case of this approach, while at the same time, this allows us to formulate robustness considerations for discrete-time CNNS.

We propose an extension of Convolutional Neural Networks (CNNs) to graph-structured data, including strided convolutions and data augmentation on graphs. Our method matches the accuracy of state-of-the-art CNNs when applied on images, without any prior about their 2D regular structure. On fMRI data, we obtain a significant gain in accuracy compared with existing graph-based alternatives.

Recent years have witnessed an extensive popularity of convolutional neural networks (CNNs) in various computer vision and artificial intelligence applications. However, the performance gains have come at a cost of substantially intensive computation complexity, which prohibits its usage in resource-limited applications like mobile or embedded devices. While increasing attention has been paid t...

Region-based Convolutional Neural Networks (R-CNNs) have achieved great success in the field of object detection. The existing R-CNNs usually divide a Region-of-Interest (ROI) into grids, and then localize objects by utilizing the spatial information reflected by the relative position of each grid in the ROI. In this paper, we propose a novel featureencoding approach, where spatial information ...

a novel neuro-based method is introduced to solve the laminar boundary layer and the turbulent free jet equations. the proposed method is based on cellular neural networks, cnns, which are recently applied widely to solve partial differential equations. the effectiveness of the method is illustrated through some examples.

Whereas CNNs have demonstrated immense progress in many vision problems, they suffer from a dependence on monumental amounts of labeled training data. On the other hand, dictionary learning does not scale to the size of problems that CNNs can handle, despite being very effective at low-level vision tasks such as denoising and inpainting. Recently, interest has grown in adapting dictionary learn...

Convolutional neural networks (CNNs) have become effective instruments in facial expression recognition. Very good results can be achieved with deep CNNs possessing many layers and providing a good internal representation of the learned data. Due to the potentially high complexity of CNNs on the other hand they are prone to overfitting and as a result, regularization techniques are needed to im...

We address the problem of object recognition from RGB-D images using deep convolutional neural networks (CNNs). We advocate the use of 3D CNNs to fully exploit the 3D spatial information in depth images as well as the use of pretrained 2D CNNs to learn features from RGB-D images. There exists currently no large scale dataset available comprising depth information as compared to those for RGB da...

Convoulutional Neural Networks (CNNs) exhibit extraordinary performance on a variety of machine learning tasks. However, their mathematical properties and behavior are quite poorly understood. There is some work, in the form of a framework, for analyzing the operations that they perform. The goal of this project is to present key results from this theory, and provide intuition for why CNNs work.

We show that a deep convolutional network with an architecture inspired by the models used in image recognition can yield accuracy similar to a long-short term memory (LSTM) network, which achieves the state-of-the-art performance on the standard Switchboard automatic speech recognition task. Moreover, we demonstrate that merging the knowledge in the CNN and LSTM models via model compression fu...