Application of Traditional Cultural Symbols in Art Design under the Background of Artificial Intelligence

In order to solve the declining influence of traditional cultural symbols, the research on traditional cultural symbols has become more meaningful. This article aims to study the application of traditional cultural symbols in art design under the background of artificial intelligence. In this paper, a fractal model with self-combined nonlinear function changes is constructed. By combining nonlinear transformations and multiparameter adjustments, various types of fractal models can be automatically rendered. The convolutional neural network algorithm is used to extract the characteristics of the style picture, and it is compared with the trained picture many times to avoid the problem of excessive tendency of the image with improper weight. The improved L-BFGS algorithm is also used to optimize the loss of the traditional L-BFGS, which improves the quality of the generated pictures and reduces the noise of the chessboard. The experimental results in this paper show that the improved L-BFGS algorithm has the least loss and the shortest time in the time used for more than 500 s. Compared with the traditional AdaGrad method, its loss is reduced by about 62%; compared with the traditional AdaDelta method, its loss is reduced by 46%. Its loss is reduced by about 8% compared with the newly optimized Adam method, which is a great improvement.

DYNAMIC SUBSTITUTION BOXES OF SYMMETRIC BLOCK ENCRYPTION ALGORITHMS

The paper deals with nonlinear transformations of well-known symmetric block algorithms such as AES, Kuznyechik, SM4, BelT, and Kalyna. A brief description of the substitution boxes for these algorithms is given. The properties of nonlinearity of the described substitution boxes are investigated with the calculation of the corresponding values. Based on the property of nonlinearity, a method for generating a dynamic substitution box is proposed. The purpose of this method is to generate dynamic substitution boxes (S-boxes) that change depending on the values of some parameter obtained from the secret key of the algorithm. Considering that linear and differential cryptanalysis uses known substitution boxes, the main advantage of the new method is that S-boxes are randomly key-dependent and unknown. Experiments were also carried out to implement this method. The resulting dynamic substitution boxes were tested for nonlinearity and the results were compared with the original nonlinearity values of the same substitution boxes.

Using Perfusion Contrast for Spatial Normalization of ASL MRI Images in a Pediatric Craniosynostosis Population

Spatial normalization is an important step for group image processing and evaluation of mean brain perfusion in anatomical regions using arterial spin labeling (ASL) MRI and is typically performed via high-resolution structural brain scans. However, structural segmentation and/or spatial normalization to standard space is complicated when gray-white matter contrast in structural images is low due to ongoing myelination in newborns and infants. This problem is of particularly clinical relevance for imaging infants with inborn or acquired disorders that impair normal brain development. We investigated whether the ASL MRI perfusion contrast is a viable alternative for spatial normalization, using a pseudo-continuous ASL acquired using a 1.5 T MRI unit (GE Healthcare). Four approaches have been compared: (1) using the structural image contrast, or perfusion contrast with (2) rigid, (3) affine, and (4) nonlinear transformations – in 16 healthy controls [median age 0.83 years, inter-quartile range (IQR) ± 0.56] and 36 trigonocephaly patients (median age 0.50 years, IQR ± 0.30) – a non-syndromic type of craniosynostosis. Performance was compared quantitatively using the real-valued Tanimoto coefficient (TC), visually by three blinded readers, and eventually by the impact on regional cerebral blood flow (CBF) values. For both patients and controls, nonlinear registration using perfusion contrast showed the highest TC, at 17.51 (CI 6.66–49.38) times more likely to have a higher rating and 17.45–18.88 ml/100 g/min higher CBF compared with the standard normalization. Using perfusion-based contrast improved spatial normalization compared with the use of structural images, significantly affected the regional CBF, and may open up new possibilities for future large pediatric ASL brain studies.

The predictive skill of convolutional neural networks models for disease forecasting

In this paper we investigate the utility of one-dimensional convolutional neural network (CNN) models in epidemiological forecasting. Deep learning models, in particular variants of recurrent neural networks (RNNs) have been studied for ILI (Influenza-Like Illness) forecasting, and have achieved a higher forecasting skill compared to conventional models such as ARIMA. In this study, we adapt two neural networks that employ one-dimensional temporal convolutional layers as a primary building block—temporal convolutional networks and simple neural attentive meta-learners—for epidemiological forecasting. We then test them with influenza data from the US collected over 2010-2019. We find that epidemiological forecasting with CNNs is feasible, and their forecasting skill is comparable to, and at times, superior to, plain RNNs. Thus CNNs and RNNs bring the power of nonlinear transformations to purely data-driven epidemiological models, a capability that heretofore has been limited to more elaborate mechanistic/compartmental disease models.

DeepLearning for Computer Vision Problems

Deep learning is a combined area between neural network and machine learning. Over the last years, deep learning methods have been shown to outperform previous state-of-the-art machine learning techniques in several fields. With computer vision being one of the most prominent cases, the deep learning methodology applies nonlinear transformations and model abstractions of high levels in large databases. Therefore, an overview of DL methodology is provided along with its major modal principals and its hierarchy, which are presented and compared with the more conventional algorithms. Likewise, its popularity and usefulness in the artificial intelligence world are discussed, and some important techniques that increase DL performance are highlighted.

Robust Learning with Implicit Residual Networks

In this effort, we propose a new deep architecture utilizing residual blocks inspired by implicit discretization schemes. As opposed to the standard feed-forward networks, the outputs of the proposed implicit residual blocks are defined as the fixed points of the appropriately chosen nonlinear transformations. We show that this choice leads to the improved stability of both forward and backward propagations, has a favorable impact on the generalization power, and allows for control the robustness of the network with only a few hyperparameters. In addition, the proposed reformulation of ResNet does not introduce new parameters and can potentially lead to a reduction in the number of required layers due to improved forward stability. Finally, we derive the memory-efficient training algorithm, propose a stochastic regularization technique, and provide numerical results in support of our findings.