Internet application providers now have more incentive than ever to collect user data, which greatly increases the risk of user privacy violations due to the emerging of deep neural networks. In this paper, we propose TensorClog-a poisoning attack technique that is designed for privacy protection against deep neural networks. TensorClog has three properties with each of them serving a privacy protection purpose: 1) training on TensorClog poisoned data results in lower inference accuracy, reducing the incentive of abusive data collection; 2) training on TensorClog poisoned data converges to a larger loss, which prevents the neural network from learning the privacy; and 3) TensorClog regularizes the perturbation to remain a high structure similarity, so that the poisoning does not affect the actual content in the data. Applying our TensorClog poisoning technique to CIFAR-10 dataset results in an increase in both converged training loss and test error by 300% and 272%, respectively. It manages to maintain data's human perception with a high SSIM index of 0.9905. More experiments including different limited information attack scenarios and a real-world application transferred from pre-trained ImageNet models are presented to further evaluate TensorClog's effectiveness in more complex situations. INDEX TERMS Deep neural networks, poisoning attack, privacy, adversarial attack.
Tea polysaccharides exhibit importantly multiple bioactivities, but a very few of them could be absorbed through the small intestine. To enhance the absorption efficacy of tea polysaccharides, a cationic vitamin...
a b s t r a c tThe analysis of cerebrovascular shape is important for the diagnose and pathologic identification. But as the limitation of the segmentation algorithm, the complete cerebrovascular volume data are difficult to obtain. So the triangle mesh of the vessel model generated for the medical images may appear many gaps. In the paper, we present a extension algorithm for Ball B-Spline curve with G 2 continuity to repair the cerebrovascular structure from time-of-flight (TOF) magnetic resonance angiography (MRA) data. Ball B-Spline curve has its distinct advantages in representing a 3D tube like organs. A ball Bezier segment is used to construct the extending part and G 2 -continuity is applied to describe the smoothness at the joints. Fairness of the extending ball Bezier curve segment is achieved by minimizing energy objective functions for the center curve and the radius function separately. New control balls are computed by unclamping algorithm to represent the whole extended ball B-Spline curve. The experimental results demonstrate the effectiveness of our algorithm. The final results show that the proposed method provides good blending result, especially for those blood vessels of small size.
Spiking neural network (SNN) that converted from conventional deep neural network (DNN) has shown great potential as a solution for fast and efficient recognition. A layer-wise quantisation method based on retraining is proposed to quantise the activation of DNN, which reduces the number of time steps required by converted SNN to achieve minimal accuracy loss. Pooling function is incorporated into convolutional layers to reduce at most 20% of spiking neurons. The converted SNNs achieved 99.15% accuracy on MNIST and 82.9% on CIFAR10 by only seven time steps, and only 10-40% of spikes need to be processed compared with networks using traditional algorithms. The experimental results show that the proposed methods are able to build hardware-friendly SNNs with ultra-low-inference latency.
Privacy recently emerges as a severe concern in deep learning, that is, sensitive data must be prohibited from being shared with the third party during deep neural network development. In this paper, we propose Morphed Learning (MoLe), an efficient and secure scheme to deliver deep learning data. MoLe has two main components: data morphing and Augmented Convolutional (Aug-Conv) layer. Data morphing allows data providers to send morphed data without privacy information, while Aug-Conv layer helps deep learning developers to apply their networks on the morphed data without performance penalty. MoLe provides stronger security while introducing lower overhead compared to GAZELLE (USENIX Security 2018), which is another method with no performance penalty on the neural network. When using MoLe for VGG-16 network on CIFAR dataset, the computational overhead is only 9% and the data transmission overhead is 5.12%. As a comparison, GAZELLE has computational overhead of 10,000 times and data transmission overhead of 421,000 times. In this setting, the attack success rate of adversary is 7.9 × 10 −90 for MoLe and 2.9 × 10 −30 for GAZELLE, respectively.
In order to reach zero energy consumption of the current building stock, adding a new insulating envelope with Renewable Energy Sources onto the existing building is necessary. This can be achieved by using prefabricated modules, automation and robotics. Since the topic is complex, three main subcategories were defined: Data Flow, Off-site Manufacturing, and On-site Installation. Latest studies suggest that there are still gaps in the way of achieving economically feasible solutions. In other words, there must be a reduction in the working time achieved in each sub-category. In this paper four different solutions are explained: 1) online data processing of the building, 2) automated layout definition, 3) accurate measurement with targets and 4) Automated CAM generation and Manufacturing. The solutions are still being developed but the current results are promising.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.