Faking Signals to Fool Deep Neural Networks in AMC via Few Data Points

Dropout regularization has been widely used in various deep neural networks to combat overfitting. It works by training a network to be more robust on information-degraded data points for better generalization. Conventional dropout and variants are often applied to individual hidden units in a layer to break up co-adaptations of feature detectors. In this paper, we propose an adaptive dropout to reduce the co-adaptations in a group-wise manner by coarse semantic information to improve feature discriminability. In particular, we showed that adjusting the dropout probability based on local feature densities can not only improve the classification performance significantly but also enhance the network robustness against adversarial examples in some cases. The proposed approach was evaluated in comparison with the baseline and several state-of-the-art adaptive dropouts over four public datasets of Fashion-MNIST, CIFAR-10, CIFAR-100 and SVHN.

Download Full-text

AI for conservation

XRDS Crossroads The ACM Magazine for Students ◽

10.1145/3466859 ◽

2021 ◽

Vol 27 (4) ◽

pp. 26-29

Author(s):

Zezhou Cheng ◽

Subhransu Maji ◽

Daniel Sheldon

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Bird Migration ◽

Weather Radar ◽

Radar Data ◽

Continental Scale ◽

Data Points

How deep neural networks can process millions of weather radar data points to help researchers monitor continental-scale bird migration.

Download Full-text

Time-Effective Simulation Methodology for Broadband Achromatic Metalens Using Deep Neural Networks

Nanomaterials ◽

10.3390/nano11081966 ◽

2021 ◽

Vol 11 (8) ◽

pp. 1966

Author(s):

Chun-Yuan Fan ◽

Guo-Dung J. Su

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Calculated Data ◽

Visible Spectrum ◽

Electromagnetic Simulation ◽

Simulation Methodology ◽

Electromagnetic Problems ◽

Huge Data ◽

Data Points ◽

Data Library

Metasurface has demonstrated potential and novel optical properties in previous research. The prevailing method of designing a macroscale metasurface is based on the local periodic approximation. Such a method relies on the pre-calculated data library, including phase delay and transmittance of the nanostructure, which is rigorously calculated by the electromagnetic simulation. However, it is usually time-consuming to design a complex metasurface such as broadband achromatic metalens due the required huge data library. This paper combined different numbers of nanofins and used deep neural networks to train our data library, and the well-trained model predicted approximately ten times more data points, which show a higher transmission for designing a broadband achromatic metalens. The results showed that the focusing efficiency of designed metalens using the augmented library is up to 45%, which is higher than that using the original library over the visible spectrum. We demonstrated that the proposed method is time-effective and accurate enough to design complex electromagnetic problems.

Download Full-text

Bird Detection using Siamese Neural Network

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.e2468.059720 ◽

2020 ◽

Vol 9 (7) ◽

pp. 1168-1171

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Neural Networks ◽

Bird Species ◽

Siamese Network ◽

The Neural Network ◽

Data Points ◽

Bird Detection ◽

Processing Techniques ◽

Do So

The proposed system uses deep neural networks for identifying bird species. The model will be trained on bird images that are coming in the endangered species category. The application can also handle new data points, unlike existing systems that require model re-training for accommodating new data. The system can identify bird species in a large view of the image. The model will be trained using a convolutional neural network-based architecture called Siamese Network. This network is also called one-shot learning which means that it requires only few training example for each class. Existing models use image processing techniques or vanilla convolutional neural networks for classifying bird images. These models cannot accommodate new images and have to be retrained to do so. There is no commercially available system that can detect a species of bird in high resolution / large image. While in the Siamese network we only have to add new data, there is no need to retraining the neural network.

Download Full-text

Modeling Deep Neural Networks to Learn Maintenance and Repair Costs of Educational Facilities

Buildings ◽

10.3390/buildings11040165 ◽

2021 ◽

Vol 11 (4) ◽

pp. 165

Author(s):

Jimyong Kim ◽

Sangguk Yum ◽

Seunghyun Son ◽

Kiyoung Son ◽

Junseo Bae

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Holistic Approach ◽

Absolute Error ◽

Facilities Management ◽

Educational Facilities ◽

Modeling Framework ◽

Maintenance And Repair ◽

Repair Costs ◽

Data Points

Educational facilities hold a higher degree of uncertainty in predicting maintenance and repair costs than other types of facilities. Moreover, achieving accurate and reliable maintenance and repair costs is essential, yet very little is known about a holistic approach to learning them by incorporating multi-contextual factors that affect maintenance and repair costs. This study fills this knowledge gap by modeling and validating deep neural networks to efficiently and accurately learn maintenance and repair costs, drawing on 1213 high-confidence data points. The developed model learns and generalizes claim payout records on the maintenance and repair costs from sets of facility asset information, geographic profiles, natural hazard records, and other causes of financial losses. The robustness of the developed model was tested and validated by measuring the root mean square error and mean absolute error values. This study attempted to propose an analytical modeling framework that can accurately learn various factors, significantly affecting the maintenance and repair costs of educational facilities. The proposed approach can contribute to the existing body of knowledge, serving as a reference for the facilities management of other functional types of facilities.

Download Full-text