Entropy最新文献

This paper focuses on the Low-Density Algebra-Check (LDAC) code, a novel low-rate channel code derived from the Low-Density Parity-Check (LDPC) code with expanded algebra-check constraints. A method for optimizing LDAC code design using Extrinsic Information Transfer (EXIT) charts is presented. Firstly, an iterative decoding model for LDAC is established according to its structure, and a method for plotting EXIT curves of the algebra-check node decoder is proposed. Then, the performance of two types of algebra-check nodes under different conditions is analyzed via EXIT curves. Finally, a low-rate LDAC code with enhanced coding gain is constructed, demonstrating the effectiveness of the proposed method.

The efficacy of machine learning has increased exponentially over the past decade. The utilization of machine learning to predict and design materials has become a pivotal tool for accelerating materials development. High-entropy alloys are particularly intriguing candidates for exemplifying the potency of machine learning due to their superior mechanical properties, vast compositional space, and intricate chemical interactions. This review examines the general process of developing machine learning models. The advances and new algorithms of machine learning in the field of high-entropy alloys are presented in each part of the process. These advances are based on both improvements in computer algorithms and physical representations that focus on the unique ordering properties of high-entropy alloys. We also show the results of generative models, data augmentation, and transfer learning in high-entropy alloys and conclude with a summary of the challenges still faced in machine learning high-entropy alloys today.

This paper presents a method for lossless compression of images with fast decoding time and the option to select encoder parameters for individual image characteristics to increase compression efficiency. The data modeling stage was based on linear and nonlinear prediction, which was complemented by a simple block for removing the context-dependent constant component. The prediction was based on the Iterative Reweighted Least Squares (IRLS) method which allowed the minimization of mean absolute error. Two-stage compression was used to encode prediction errors: an adaptive Golomb and a binary arithmetic coding. High compression efficiency was achieved by using an author's context-switching algorithm, which allows several prediction models tailored to the individual characteristics of each image area. In addition, an analysis of the impact of individual encoder parameters on efficiency and encoding time was conducted, and the efficiency of the proposed solution was shown against competing solutions, showing a 9.1% improvement in the bit average of files for the entire test base compared to JPEG-LS.

Diagnosing faults in wheelset bearings is critical for train safety. The main challenge is that only a limited amount of fault sample data can be obtained during high-speed train operations. This scarcity of samples impacts the training and accuracy of deep learning models for wheelset bearing fault diagnosis. Studies show that the Auxiliary Classifier Generative Adversarial Network (ACGAN) demonstrates promising performance in addressing this issue. However, existing ACGAN models have drawbacks such as complexity, high computational expenses, mode collapse, and vanishing gradients. Aiming to address these issues, this paper presents the Transformer and Auxiliary Classifier Generative Adversarial Network (TACGAN), which increases the diversity, complexity and entropy of generated samples, and maximizes the entropy of the generated samples. The transformer network replaces traditional convolutional neural networks (CNNs), avoiding iterative and convolutional structures, thereby reducing computational expenses. Moreover, an independent classifier is integrated to prevent the coupling problem, where the discriminator is simultaneously identified and classified in the ACGAN. Finally, the Wasserstein distance is employed in the loss function to mitigate mode collapse and vanishing gradients. Experimental results using the train wheelset bearing datasets demonstrate the accuracy and effectiveness of the TACGAN.

The transient plane source (TPS) method heat transfer model was established. A body-fitted coordinate system is proposed to transform the unstructured grid structure to improve the speed of solving the heat transfer direct problem of the winding probe. A parallel Bayesian optimization algorithm based on a multi-objective hybrid strategy (MHS) is proposed based on an inverse problem. The efficiency of the thermophysical properties inversion was improved. The results show that the meshing method of 30° is the best. The transformation of body-fitted mesh is related to the orthogonality and density of the mesh. Compared with parameter inversion the computational fluid dynamics (CFD) software, the absolute values of the relative deviations of different materials are less than 0.03%. The calculation speeds of the body-fitted grid program are more than 36% and 91% higher than those of the CFD and self-developed unstructured mesh programs, respectively. The application of body-fitted coordinate system effectively improves the calculation speed of the TPS method. The MHS is more competitive than other algorithms in parallel mode, both in terms of accuracy and speed. The accuracy of the inversion is less affected by the number of initial samples, time range, and parallel points. The number of parallel points increased from 2 to 6, reducing the computation time by 66.6%. Adding parallel points effectively accelerates the convergence of algorithms.

With the rapid increase in global data and rapid development of information technology, DNA sequences have been collected and manipulated on computers. This has yielded a new and attractive field of bioinformatics, DNA storage, where DNA has been considered as a great potential storage medium. It is known that one gram of DNA can store 215 GB of data, and the data stored in the DNA can be preserved for tens of thousands of years. In this study, a lossless and reversible DNA data storage method was proposed. The proposed approach employs a vector representation of each DNA base in a two-dimensional (2D) spatial domain for both encoding and decoding. The structure of the proposed method is reversible, rendering the decompression procedure possible. Experiments were performed to investigate the capacity, compression ratio, stability, and reliability. The obtained results show that the proposed method is much more efficient in terms of capacity than other known algorithms in the literature.

In today's digital age, ensuring the appropriateness of content for children is crucial for their cognitive and emotional development. The rise of automated text generation technologies, such as Large Language Models like LLaMA, Mistral, and Zephyr, has created a pressing need for effective tools to filter and classify suitable content. However, the existing methods often fail to effectively address the intricate details and unique characteristics of children's literature. This study aims to bridge this gap by developing a robust framework that utilizes fine-tuned language models, classification techniques, and contextual story generation to generate and classify children's stories based on their suitability. Employing a combination of fine-tuning techniques on models such as LLaMA, Mistral, and Zephyr, alongside a BERT-based classifier, we evaluated the generated stories against established metrics like ROUGE, METEOR, and BERT Scores. The fine-tuned Mistral-7B model achieved a ROUGE-1 score of 0.4785, significantly higher than the base model's 0.3185, while Zephyr-7B-Beta achieved a METEOR score of 0.4154 compared to its base counterpart's score of 0.3602. The results indicated that the fine-tuned models outperformed base models, generating content more aligned with human standards. Moreover, the BERT Classifier exhibited high precision (0.95) and recall (0.97) for identifying unsuitable content, further enhancing the reliability of content classification. These findings highlight the potential of advanced language models in generating age-appropriate stories and enhancing content moderation strategies. This research has broader implications for educational technology, content curation, and parental control systems, offering a scalable approach to ensuring children's exposure to safe and enriching narratives.

The probability of gas escapes from steel pipelines due to different types of corrosion is studied with real failure data from an urban gas distribution network. Both the design and maintenance of the network are considered, identifying and estimating (in a Bayesian framework) an elementary multinomial model in the first case, and a more sophisticated non-homogeneous Poisson process in the second case. Special attention is paid to the elicitation of the experts' opinions. We conclude that the corrosion process behaves quite differently depending on the type of corrosion, and that, in most cases, cathodically protected pipes should be installed.

This paper proposes a novel polar coding scheme tailored for indoor visible light communication (VLC) systems. Simulation results demonstrate a significant reduction in bit error rate (BER) compared to uncoded transmission, with a coding gain of at least 5 dB. Furthermore, the reliable communication area of the VLC system is substantially extended. Building on this foundation, this study explores the joint design of polar codes and physical-layer network coding (PNC) for VLC systems. Simulation results illustrate that the BER of our scheme closely approaches that of the conventional VLC relay scheme. Moreover, our approach doubles the throughput, cuts equipment expenses in half, and boosts effective bit rates per unit time-slot twofold. This proposed design noticeably advances the performance of VLC systems and is particularly well-suited for scenarios with low-latency demands.

In recent years, the rapid growth of video data posed challenges for storage and transmission. Video compression techniques provided a viable solution to this problem. In this study, we proposed a bidirectional coding video compression model named DeepBiVC, which was based on two-stage learning. Firstly, we conducted preprocessing on the video data by segmenting the video flow into groups of continuous image frames, with each group comprising five frames. Then, in the first stage, we developed an image compression module based on an invertible neural network (INN) model to compress the first and last frames of each group. In the second stage, we designed a video compression module that compressed the intermediate frames using bidirectional optical flow estimation. Experimental results indicated that DeepBiVC outperformed other state-of-the-art video compression methods regarding PSNR and MS-SSIM metrics. Specifically, on the VUG dataset at bpp = 0.3, DeepBiVC achieved a PSNR of 37.16 and an MS-SSIM of 0.98.