Chained Dual-Generative Adversarial Network: A Generalized Defense Against Adversarial Attacks

Neural networks play a significant role in the field of image classification. When an input image is modified by adversarial attacks, the changes are imperceptible to the human eye, but it still leads to misclassification of the images. Researchers have demonstrated these attacks to make production self-driving cars misclassify Stop Road signs as 45 Miles Per Hour (MPH) road signs and a turtle being misclassified as AK47. Three primary types of defense approaches exist which can safeguard against such attacks i.e., Gradient Masking, Robust Optimization, and Adversarial Example Detection. Very few approaches use Generative Adversarial Networks (GAN) for Defense against Adversarial Attacks. In this paper, we create a new approach to defend against adversarial attacks, dubbed Chained Dual-Generative Adversarial Network (CD-GAN) that tackles the defense against adversarial attacks by minimizing the perturbations of the adversarial image using iterative oversampling and undersampling using GANs. CD-GAN is created using two GANs, i.e., CDGAN’s Sub-Resolution GAN and CDGAN’s Super-Resolution GAN. The first is CDGAN’s Sub-Resolution GAN which takes the original resolution input image and oversamples it to generate a lower resolution neutralized image. The second is CDGAN’s Super-Resolution GAN which takes the output of the CDGAN’s Sub-Resolution and undersamples, it to generate the higher resolution image which removes any remaining perturbations. Chained Dual GAN is formed by chaining these two GANs together. Both of these GANs are trained independently. CDGAN’s Sub-Resolution GAN is trained using higher resolution adversarial images as inputs and lower resolution neutralized images as output image examples. Hence, this GAN downscales the image while removing adversarial attack noise. CDGAN’s Super-Resolution GAN is trained using lower resolution adversarial images as inputs and higher resolution neutralized images as output images. Because of this, it acts as an Upscaling GAN while removing the adversarial attak noise. Furthermore, CD-GAN has a modular design such that it can be pre-fixed to any existing classifier without any retraining or extra effort, and can defend any classifier model against adversarial attack. In this way, it is a Generalized Defense against adversarial attacks, capable of defending any classifier model against any attacks. This enables the user to directly integrate CD-GAN with an existing production deployed classifier smoothly. CD-GAN iteratively removes the adversarial noise using a multi-step approach in a modular approach. It performs comparably to the state of the arts with mean accuracy of 33.67 while using minimal compute resources in training.     

Visual learning by coevolutionary feature synthesis.

In this paper, a novel genetically inspired visual learning method is proposed. Given the training raster images, this general approach induces a sophisticated feature-based recognition system. It employs the paradigm of cooperative coevolution to handle the computational difficulty of this task. To represent the feature extraction agents, the linear genetic programming is used. The paper describes the learning algorithm and provides a firm rationale for its design. Different architectures of recognition systems are considered that employ the proposed feature synthesis method. An extensive experimental evaluation on the demanding real-world task of object recognition in synthetic aperture radar (SAR) imagery shows the ability of the proposed approach to attain high recognition performance in different operating conditions.     

Learning to Perceive Objects for Autonomous Navigation

Current machine perception techniques that typically use segmentation followed by object recognition lack the required robustness to cope with the large variety of situations encountered in real-world navigation. Many existing techniques are brittle in the sense that even minor changes in the expected task environment (e.g., different lighting conditions, geometrical distortion, etc.) can severely degrade the performance of the system or even make it fail completely. In this paper we present a system that achieves robust performance by using local reinforcement learning to induce a highly adaptive mapping from input images to segmentation strategies for successful recognition. This is accomplished by using the confidence level of model matching as reinforcement to drive learning. Local reinforcement learning gives rises to better improvement in recognition performance. The system is verified through experiments on a large set of real images of traffic signs.     

Crystallographically selective nanopatterning of graphene on SiO<Subscript>2</Subscript>

Graphene has many advantageous properties, but its lack of an electronic band gap makes this two-dimensional material impractical for many nanoelectronic applications, for example, field-effect transistors. This problem can be circumvented by opening up a confinement-induced gap, through the patterning of graphene into ribbons having widths of a few nanometres. The electronic properties of such ribbons depend on both their size and the crystallographic orientation of the ribbon edges. Therefore, etching processes that are able to differentiate between the zigzag and armchair type edge terminations of graphene are highly sought after. In this contribution we show that such an anisotropic, dry etching reaction is possible and we use it to obtain graphene ribbons with zigzag edges. We demonstrate that the starting positions for the carbon removal reaction can be tailored at will with precision.     

Effect of interesterified dietary fat mixtures on various lipid indices and liver metal content of rats in fat-metabolism disturbance

Adult male and female Wistar rats were fed isoenergetically for a 6 week period with a lipogenic diet containing a 20% fat mixture which caused fat-metabolism disturbance. One group consumed the mixture of sunflower oil and lard in a ratio of 0.91 P/S, the other group the interesterified form of the same mixture. The linoleic acid content of the mixtures was 0.4%. The fat mixture of the third group's diet was adjusted with soya oil to a linoleic-acid content of 0.8%. The results were compared to the control data obtained in rats fed with a normal diet. The changes in various lipid indices of the serum and the liver and the levels of some metals in the liver were analysed. It was found that; in comparison to the control on the effect of the lipogenic diet the total lipid and cholesterol contents of the serum increased significantly in all groups (a significantly higher value being observed for the females than for the males) whereas the HDL-C content decreased.(ABSTRACT TRUNCATED AT 250 WORDS)