Deep learning has recently achieved best-in-class performance in several fields, including biomedical domains such as X-ray images. Yet, data scarcity poses a strict limit on training successful deep learning systems ...
详细信息
Deep learning has recently achieved best-in-class performance in several fields, including biomedical domains such as X-ray images. Yet, data scarcity poses a strict limit on training successful deep learning systems in many, if not most, biomedical applications, including those involving brain images. In this study, we translate state-of-the-art transfer learning techniques for single-subject prediction of simpler (sex and age) and more complex phenotypes (number of people in household, household income, fluid intelligence and smoking behavior). We fine-tuned 2D and 3D ResNet-18 convolutionalneuralnetworks for target phenotype predictions from brain images of similar to 40,000 UK Biobank participants, after pretraining on YouTube videos from the Kinetics dataset and natural images from the ImageNet dataset. Transfer learning was effective on several phenotypes, especially sex and age classification. Additionally, transfer learning in particular outperformed deep learning models trained from scratch especially on smaller sample sizes. The out-of-sample performance using transfer learning from previously learned knowledge based on real-world images and videos could unlock the potential in many areas of imaging neuroscience where deep learning solutions are currently infeasible. (c) 2022 Elsevier Ltd. All rights reserved.
Power analysis methods are commonly used for evaluating the security of cryptographic *** are characteristically low-cost and display a high success rate and the ability to obtain important device information, e.g., k...
详细信息
Power analysis methods are commonly used for evaluating the security of cryptographic *** are characteristically low-cost and display a high success rate and the ability to obtain important device information, e.g., keys. Given the current wide application of deep-learning technology, there is a growing tendency to incorporate power-analysis technology in *** study investigates non-profiled deep-learning-based power analysis. The labels used in this attack are uncertain,and the attack conditions required are greatly reduced. We choose the Recurrent neuralnetwork(RNN), multilayer perceptron, and convolutional neural network algorithms,which use the same network structure, to recover the keys for the SM4 software and DES hardware *** propose combining the RNN algorithm with power analysis, and validate the benefits experimentally. The experimental results show that they all successfully recover the correct key for the SM4 software implementation,although the RNN algorithm by itself achieves a better effect. This conclusion also applies to attacks on the DES hardware implementation but is limited to labels based on the bit model.
Hyperspectral satellite imagery (HSI) is an advanced technology for object detection because it provides a large amount of information. Thus, the classification of HSIs is very complicated, so the methods of reducing ...
详细信息
Hyperspectral satellite imagery (HSI) is an advanced technology for object detection because it provides a large amount of information. Thus, the classification of HSIs is very complicated, so the methods of reducing spectral or spatial information generally degrade the quality of classification. In order to solve this problem and guarantee faster and more efficient processing, we propose a smart feature extraction (SFE) and classification by convolutionalneuralnetwork (2D-CNN) method made up of two parts. The first consists in reducing spectral information by a probabilistic method based on the Softmax function. The second is classification by processing batches of data in the proposed CNN network. The method was tested on two public hyperspectral images (Indian Pines and SalinasA) to prove its effectiveness in increasing classification accuracy and reducing computing time.
暂无评论