Search

Multi-view clustering considers the diversity of different views and fuses these views to produce a more accurate and robust partition than single-view clustering. It is a key problem of multi-view clustering research to allocate each view reasonably based on its contribution value. In this paper, we propose a weighted multi-view clustering model via sparse graph learning to cope with allocation of different views. The proposed idea is to assign different view weights instead of equal view weights to learn a high-quality shared similarity matrix for multi-view clustering. In our new proposed method, it can consider the clustering capacity heterogeneity of different views in fusion by assigning a weight for each view so that each view special feature are fully excavated, and improve the performance of multi-view clustering. Moreover, our proposed method can directly obtained cluster indicators by imposing low rank constraints without any post-processing operations. In addition, our model is proposed based on sparse graph, so that the outliers and noise in each view data are well handled and the robustness of the algorithm is effectively guaranteed. Finally, numerous experimental results are conducted on different sizes benchmark datasets, and show that the performance of our algorithm is quite satisfactory. The code of our proposed method is publicly available at https://github.com/zhoujie05/A-weighted-multi-view-clustering-via-sparse-graph-learning. [ABSTRACT FROM AUTHOR]

Deep neural networks are vulnerable to adversarial examples which are generated by adding carefully crafted perturbations on benign examples. Some research works explore the transferability of adversarial examples between hetero-modal models from images to videos. However, these works add adversarial perturbations to each frame of the video without considering sparse attack. To bridge this gap, we propose a label independent cross-modal transferable adversarial video attack with sparse strategy called SS-CMT, which efficiently generates sparse adversarial video examples with low perturbations and high transferability. Specifically, we propose a sparse strategy to select sparse frames from benign video examples and propose a cross-modal transferable attack to add adversarial perturbations to sparse frames. Besides, our method does not use the labels of video examples in the process of generating sparse adversarial examples. We also explore attacking an ensemble of alternative models to boost the transferability of generated adversarial examples, called ENS-SS-CMT. Extensive experiments on Kinetics-400 and UCF-101 demonstrate the excellent performance of the proposed methods SS-CMT and ENS-SS-CMT in terms of high transferability, high efficiency, and low perturbations. Among them, the proposed method ENS-SS-CMT outperforms the state-of-the-art method on overall performance. [ABSTRACT FROM AUTHOR]

A pure pair of size t in a graph G is a pair A, B of disjoint subsets of V(G), each of cardinality at least t, such that A is either complete or anticomplete to B. It is known that, for every forest H, every graph on n ≥ 2 vertices that does not contain H or its complement as an induced subgraph has a pure pair of size Ω (n) ; furthermore, this only holds when H or its complement is a forest. In this paper, we look at pure pairs of size n 1 - c , where 0 < c < 1 . Let H be a graph: does every graph on n ≥ 2 vertices that does not contain H or its complement as an induced subgraph have a pure pair of size Ω (| G | 1 - c) ? The answer is related to the congestion of H, the maximum of 1 - (| J | - 1) / | E (J) | over all subgraphs J of H with an edge. (Congestion is nonnegative, and equals zero exactly when H is a forest.) Let d be the smaller of the congestions of H and H ¯ . We show that the answer to the question above is "yes" if d ≤ c / (9 + 15 c) , and "no" if d > c . [ABSTRACT FROM AUTHOR]

The reconstruction of MR images has always been a challenging inverse problem in medical imaging. Acceleration of MR scanning is of great importance for clinical research and cutting-edge applications. One of the primary efforts to achieve this is using compressed sensing (CS) theory. The CS aims to reconstruct MR images using a small number of sampled data in k-space. The CS-MRI techniques face challenges, including the potential loss of fine structure and increased computational complexity. We introduce a novel framework based on a regularized sparse recovery problem and a sharpening step to improve the CS-MRI approaches regarding fine structure loss under high acceleration factors. This problem is solved via the Half Quadratic Splitting (HQS) approach. The inverse problem for reconstructing MR images is converted into two distinct sub-problems, each of which can be solved separately. One key feature of the proposed approach is the replacement of one sub-problem with a denoiser. This regularization assists the optimization of the Smoothed (SL0) norm in escaping local minimums and enhances its precision. The proposed method consists of smoothing, feature modification, and Smoothed cost function optimization. The proposed approach improves the SL0 algorithm for MRI reconstruction without complicating it. The convergence of the proposed approach is illustrated analytically. The experimental results show an acceptable performance of the proposed method compared to the network-based approaches. [ABSTRACT FROM AUTHOR]

Spin torque nano-oscillators possessing fast nonlinear dynamics and short-term memory functions are potentially able to achieve energy-efficient neuromorphic computing. In this study, we introduce an activation-state controllable spin neuron unit composed of vertically coupled vortex spin torque oscillators and a V – I source circuit is proposed and used to build an energy-efficient sparse reservoir computing (RC) system to solve nonlinear dynamic system prediction task. Based on micromagnetic and electronic circuit simulation, the Mackey–Glass chaotic time series and the real motor vibration signal series can be predicted by the RC system with merely 20 and 100 spin neuron units, respectively. Further study shows that the proposed sparse reservoir system could reduce energy consumption without significantly compromising performance, and a minimal response from inactivated neurons is crucial for maintaining the system's performance. The accuracy and signal processing speed show the potential of the proposed sparse RC system for high-performance and low-energy neuromorphic computing. [ABSTRACT FROM AUTHOR]

Abstract The reconstruction of MR images has always been a challenging inverse problem in medical imaging. Acceleration of MR scanning is of great importance for clinical research and cutting-edge applications. One of the primary efforts to achieve this is using compressed sensing (CS) theory. The CS aims to reconstruct MR images using a small number of sampled data in k-space. The CS-MRI techniques face challenges, including the potential loss of fine structure and increased computational complexity. We introduce a novel framework based on a regularized sparse recovery problem and a sharpening step to improve the CS-MRI approaches regarding fine structure loss under high acceleration factors. This problem is solved via the Half Quadratic Splitting (HQS) approach. The inverse problem for reconstructing MR images is converted into two distinct sub-problems, each of which can be solved separately. One key feature of the proposed approach is the replacement of one sub-problem with a denoiser. This regularization assists the optimization of the Smoothed $$\:{\ell}_{0}$$ (SL0) norm in escaping local minimums and enhances its precision. The proposed method consists of smoothing, feature modification, and Smoothed $$\:{\ell}_{0}$$ cost function optimization. The proposed approach improves the SL0 algorithm for MRI reconstruction without complicating it. The convergence of the proposed approach is illustrated analytically. The experimental results show an acceptable performance of the proposed method compared to the network-based approaches.

In this paper, we introduce a new vectorized MATLAB-based algorithm for efficient serial computation of global matrix/force arising from finite element method (FEM) for meshes of any type and approximation order in linear elasticity. Because for-loops in MATLAB are very slow, we propose a modified process that takes advantage of vectorization and sparse assembly to achieve good performance while using the same memory as the standard algorithm. For this purpose, by using good programming practices, the implementation of this scheme is succinctly described and can be integrated into any MATLAB package dealing with FEM. Specifically, attention is paid to the calculation of the triplet (row index, column index, matrix components) as well as the assembly of the global stiffness matrix, mass matrix and force vector. Additionally, an extension of the proposed approach for Mindlin plate theory and functionally graded materials is outlined. Finally, the accuracy of this strategy is verified on selected numerical tests after comparing the obtained results with those of ABAQUS. In terms of performance, the study conducted on a set of meshes considering the standard algorithm and two other well-known MATLAB vectorized algorithms revealed that: (i) for a 2D beam problem meshed with P 1 -triangle elements, a speedup of about 8 and 15 is achieved with sparse and fsparse, respectively. (ii) for a 3D plate problem meshed with P 1 -tetrahedral elements, a speedup of about 4 and 8 is achieved with sparse and fsparse, respectively. When compared to ABAQUS performance, the proposed scheme results in a computational time that is about five times smaller. [ABSTRACT FROM AUTHOR]

Structural health monitoring (SHM) ensures the safety and reliability of civil infrastructure. Autoencoders, as unsupervised learning models, offer promise for SHM by learning data features and reducing dimensionality. However, comprehensive studies comparing autoencoder models in SHM are scarce. This study investigates the effectiveness of four autoencoder-based methodologies, combined with Hotelling's T 2 statistical tool, to detect and quantify structural changes in three civil engineering structures. The methodologies are evaluated based on computational costs and their abilities to identify structural anomalies accurately. Signals from the structures, collected by accelerometers, feed the autoencoders for unsupervised classification. The latent layer values of the autoencoders are used as parameters in Hotelling's T 2 , and results are compared between classes to assess structural changes. Average execution times of each model were calculated for computational efficiency. Despite variations, computational cost did not hinder any methodology. The study demonstrates that the best fitting model, VAE- T 2 , outperforms its counterparts in identifying and quantifying structural changes. While the AE, SAE, and CAE models showed limitations in quantifying changes, they remain relevant for detecting anomalies. Continuous application and development of these techniques contribute to SHM advancements, enabling the increased safety, cost-effectiveness, and long-term durability of civil engineering structures. [ABSTRACT FROM AUTHOR]

With the advent of egocentric cameras, there are new challenges where traditional computer vision is not sufficient to handle this kind of video. Moreover, egocentric cameras often offer multiple modalities that need to be modeled jointly to exploit complimentary information. In this article, we propose a sparse low-rank bilinear score pooling approach for egocentric hand action recognition from RGB-D videos. It consists of five blocks: a baseline CNN to encode RGB and depth information for producing classification probabilities; a novel bilinear score pooling block to generate a score matrix; a sparse low-rank matrix recovery block to reduce redundant features, which is common in bilinear pooling; a one-layer CNN for frame-level classification; and an RNN for video-level classification. We proposed to fuse classification probabilities instead of traditional CNN features from RGB and depth modality, involving an effective yet simple sparse low-rank bilinear score pooling to produce a fused RGB-D score matrix. To demonstrate the efficacy of our method, we perform extensive experiments over two large-scale hand action datasets, namely, THU-READ and FPHA, and two smaller datasets, GUN-71 and HAD. We observe that the proposed method outperforms state-of-the-art methods and achieves accuracies of 78.55% and 96.87% over the THU-READ dataset in cross-subject and cross-group settings, respectively. Further, we achieved accuracies of 91.59% and 43.87% over the FPHA and Gun-71 datasets, respectively. [ABSTRACT FROM AUTHOR]

This paper investigates the potential of utilising multilevel quasi-interpolation techniques on Chebyshev sparse grids for complex numerical computations. The paper starts by laying down the motivations for choosing Chebyshev sparse grids and quasi-interpolation methods with Gaussian kernels. It delves into the practical aspects of implementing these techniques. Various numerical experiments are performed to evaluate the efficiency and limitations of the multilevel quasi-sparse interpolation methods with dimensions two dimension and three dimension. The work ultimately aims to provide a comprehensive understanding of the computational efficiency and accuracy achievable through this approach, comparing its performance with traditional methods. [ABSTRACT FROM AUTHOR]

Stochastic configuration network (SCN) is a powerful prediction model whose performance is significantly influenced by the configuration of the network parameters. To improve the prediction accuracy of the network, a cooperative stochastic configuration network (CSCN) based on a novel differential evolutionary sparrow search algorithm (DESSA), termed as DESSA-CSCN, is proposed. In the CSCN, the number of hidden layer nodes is adaptively adjusted according to the number of iterations, and the parameters of hidden nodes are cooperatively optimized by using a population-based metaheuristic algorithm. A sparse matrix is introduced to mitigate parameter overfitting caused by the increased number of hidden layer nodes. During parameter optimization, the fitness function is constructed by using the supervision mechanism of the SCN, and the DESSA is utilized as the metaheuristic algorithm to update the weights and biases. In order to verify the effectiveness of the DESSA-CSCN, several simulation experiments have been conducted. The performance of the DESSA is evaluated by the CEC2017 test suit, and the simulation results show that the DESSA exhibits better convergence accuracy and can jump out of local optima more effectively than other algorithms. The performance of the DESSA-CSCN is evaluated by 4 datasets from KEEL, and the simulation results indicate that the DESSA-CSCN achieves better prediction accuracy and faster prediction speed than other models.

Fusion of hyperspectral image (HSI) and multispectral image (MSI) is a prevalent scheme to generate a HSI with enhanced spatial resolution. Current methods often fail to sufficiently leverage the effective spectral and spatial priors existing in the observed HSI and MSI to further enhance the fusion performance. To address this limitation, this paper proposes a novel HSI-MSI fusion approach, which integrates Sparse and Low Rank with a CNN denoiser (SLRCNN) while considering spectral dictionary optimization. Firstly, an initialized spectral dictionary is derived from the HSI. Next, the spatial coefficients optimization model is established by incorporating the sparse prior, local low-rank prior, and plugged image prior simultaneously, where the l1 norm is imposed to promote the sparse prior, and the super-pixel segmentation strategy is conducted on the MSI to impose the local low-rank prior while a well-trained CNN denoiser is plugged in to enforce the image prior. Then, the spectral dictionary optimization model is constructed to refine the initial spectral dictionary, capturing more detailed spectral characteristics to further improve the fusion results. Finally, the optimization process involves applying the split-augmented Lagrangian shrinkage method and the alternating direction method of multipliers. Experimental results on simulated and real datasets, namely the Pavia University dataset, the Indian Pines dataset, and the EO-1 dataset, indicate that SLRCNN outperforms existing state-of-the-art approaches at 4x, 5x, and 6x resolutions in both qualitative and quantitative evaluation results. Specifically, the peak signal-to-noise ratio (PSNR) of SLRCNN is improved by more than 0.9 dB, 0.9 dB, and 0.2 dB while the spectral angle mapper (SAM) is decreased by more than 0.1, 0.2, and 0.2 in degree compared to other state-of-the-art methods across three datasets, respectively, which underscores the effectiveness of SLRCNN in leveraging both spatial detail reconstruction and spectral preservation.

Efficiently removing clouds from remote sensing imagery presents a significant challenge, yet it is crucial for a variety of applications. This paper introduces a novel sparse function, named the tri-fiber-wise sparse function, meticulously engineered for the targeted tasks of cloud detection and removal. This function is adept at capturing cloud characteristics across three dimensions, leveraging the sparsity of mode-1, -2, and -3 fibers simultaneously to achieve precise cloud detection. By incorporating the concept of tensor multi-rank, which describes the global correlation, we have developed a tri-fiber-wise sparse-based model that excels in both detecting and eliminating clouds from images. Furthermore, to ensure that the cloud-free information accurately matches the corresponding areas in the observed data, we have enhanced our model with an extended box-constraint strategy. The experiments showcase the notable success of the proposed method in cloud removal. This highlights its potential and utility in enhancing the accuracy of remote sensing imagery. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This paper provides a novel compressive sensing (CS) technique to compress electrocardiogram (ECG) signals. The proposed technique extends a dynamic compressed sensing method, originally based on a single lead signal, to multiple lead signals. The sensing matrix used by the dynamic sensing technique receives its components dynamically from the compressed signal. In this way, a single sensing matrix is recommended to be used for an application of many leads, for which its components are obtained from the fusion of many leads. The CS technique is tested using a variety of signals that are collected from both healthy people and those with various diseases, including bundle branch block, cardiomyopathy, and myocardial infarction. The experiment's results show that the suggested CS technique can be applied to achieve a compression ratio (CR) of 16 without affecting the metrics of the signal. [ABSTRACT FROM AUTHOR]

In ultrasonic testing of Carbon Fiber Reinforced Polymers (CFRP), signals of near-surface flaws are often submerged in interface signals, resulting in blind spots for defect detection. To address this issue, this paper presents an autocorrelation imaging algorithm that combines Multipoint Optimal Minimum Entropy Deconvolution Adjusted (MOMEDA) with ${l_0}$l0-norm sparse representation. Firstly, the ultrasonic signal can be modelled as a convolution process. By using MOMEDA, the sequence of reflected pulses in the ultrasonic signal is obtained. Then, the ${l_0}$l0-norm sparse representation is utilised to enhance the temporal resolution of the sequence, successfully separating near-surface defect signals from interface signals. Finally, the processed data is input into the autocorrelation algorithm, achieving automated imaging of near-surface flaws. Simulation results demonstrate the efficacy of the algorithm in separating overlapping signals. Finally, experimental validation was conducted on flaws at three different depths. The algorithm presented in this paper is capable of identifying all near-surface flaws, with a defect size error rate consistently below 5%. [ABSTRACT FROM AUTHOR]

Two approximation algorithms are proposed for ℓ 1 -regularized sparse rank-1 approximation to higher-order tensors. The algorithms are based on multilinear relaxation and sparsification, which are easily implemented and well scalable. In particular, the second one scales linearly with the size of the input tensor. Based on a careful estimation of the ℓ 1 -regularized sparsification, theoretical approximation lower bounds are derived. Our theoretical results also suggest an explicit way of choosing the regularization parameters. Numerical examples are provided to verify the proposed algorithms. [ABSTRACT FROM AUTHOR]

This article studies the mixed noise removal problem for hyperspectral images (HSIs), which often suffer from Gaussian noise and sparse noise. Conventional denoising models mainly employ the $L_{1}$-norm-based regularizers to remove sparse noise and ensure piecewise smoothness. However, the denoising performance is poor for highly structured images with severe noise since the $L_{1}$-norm overpenalizes large entries. To tackle this limitation, we propose a denoising model that combines tensor decomposition with two kinds of $L_{0}$-norm-based regularizers. First, we use low-rank Tucker decomposition with the Stiefel manifold to characterize the global correlation of HSIs. Then, we utilize the $L_{0}$-norm to leverage the intrinsic sparsity information of the corruption domain, thereby enhancing the effectiveness of sparse noise removal. Simultaneously, we introduce a weighted $L_{0}$-norm regularizer on the gradient of each pixel to promote the local spectral-spatial smoothness. To solve the proposed model, we design a hard-thresholding-based alternating direction method of multipliers algorithm. Instead of spending time to find a proper rank in advance, we adopt a rank-increasing scheme to dynamically adjust the tensor rank during the optimization procedure. In this way, our algorithm avoids the rank selection burden and improves computational efficiency. Finally, we test the proposed method on both synthetic and real datasets. Numerical results demonstrate its superiority, especially, the improvements of our method over the best-compared results up to 2.07 dB for mixed noise removal.

Representation-based Classification (RC) algorithms have been extensively applied in the domain of face recognition. This paper introduces a novel approach by incorporating arbitrary norms into both the objective and constraint functions of conventional RC algorithms, resulting in the development of the Generalized RC (GRC) algorithm. The primary goal of this enhancement is to fully leverage the unique advantages offered by various norms. Within the majorization-minimization framework, an iterative procedure was designed to solve the optimization problem of GRC. This approach ensures that a closed-form solution is obtained in each iteration and a locally optimal solution can be found upon convergence. Experimental results on four benchmark face databases demonstrated that GRC generally outperforms competing RC algorithms in face recognition. Data and source code of this study are publicly available on the GitHub repository at https://github.com/yuzhounh/GRC. [ABSTRACT FROM AUTHOR]

We have ported the numerical factorization and triangular solve phases of the sparse direct solver STRUMPACK to GPU. STRUMPACK implements sparse LU factorization using the multifrontal algorithm, which performs most of its operations in dense linear algebra operations on so-called frontal matrices of various sizes. Our GPU implementation off-loads these dense linear algebra operations, as well as the sparse scatter–gather operations between frontal matrices. For the larger frontal matrices, our GPU implementation relies on vendor libraries such as cuBLAS and cuSOLVER for NVIDIA GPUs and rocBLAS and rocSOLVER for AMD GPUs. For the smaller frontal matrices we developed custom CUDA and HIP kernels to reduce kernel launch overhead. Overall, high performance is achieved by identifying submatrix factorizations corresponding to sub-trees of the multifrontal assembly tree which fit entirely in GPU memory. The multi-GPU setting uses SLATE (Software for Linear Algebra Targeting Exascale) as a modern GPU-aware replacement for ScaLAPACK. On 4 nodes of SUMMIT the code runs ∼10× faster when using all 24 V100 GPUs compared to when it only uses the 168 POWER9 cores. On 8 SUMMIT nodes, using 48 V100 GPUs, the sparse solver reaches over 50TFlop/s. Compared to SuperLU, on a single V100, for a set of 17 matrices our implementation is faster for all but one matrix, and is on average 5× (median 4×) faster

The use of edge computing devices in bank outlets for passenger flow analysis, security protection, risk prevention and control is increasingly widespread, among which the performance and power consumption of AI reasoning chips have become a very important factor in the selection of edge computing devices.Aiming at the problems of recurrent neural network, such as high power consumption, weak reasoning performance and low energy efficiency, which were caused by data dependence and low data reusability, this paper realized a sparse RNN low-power accelerator with scalable voltage by using FPGA, and verifies it on the edge design and calculation equipment.Firstly, the sparse -RNN was analyzed and the processing array was designed by network compression.Secondly, due to the unbalanced workload of sparse RNN, it introduced voltage scaling method to maintain low power consumption and high throughput.Experiments show that this method could significantly improve the RNN reasoning speed of the system and reduce the processing power consumption of the chip.

Abstract In 5G communications, the acquisition of accurate channel state information (CSI) is of great importance to the hybrid beamforming of millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) system. In classical mmWave MIMO channel estimation methods, the exploitation of inherent sparse or low-rank structures has demonstrated to improve the performance. However, most high-accurate CSI estimators incur a high computational complexity and require the prior channel information, which hence present the major challenges in the practical deployment. In this work, we leverage machine learning to design the low-complexity and high-performance channel estimator. To be specific, we first formulate the CSI estimation, in the case of sparse structure, as one classical least absolute shrinkage and selection operator problem. In order to reduce the time complexity of existing compressed sensing (CS) methods, we then approximate the original optimization problem to another one, by imposing the other low-rank constraint that was barely considered by CS. We thus solve this new approximated problem and attain the near-optimal solution of the original problem. One new method excludes any prior channel information, and greatly improves the estimation performance, which only incurs a low time complexity. Simulation results demonstrate the superiority of our proposed method both in the estimation accuracy and time complexity.

Structural health monitoring (SHM) ensures the safety and reliability of civil infrastructure. Autoencoders, as unsupervised learning models, offer promise for SHM by learning data features and reducing dimensionality. However, comprehensive studies comparing autoencoder models in SHM are scarce. This study investigates the effectiveness of four autoencoder-based methodologies, combined with Hotelling’s T2 statistical tool, to detect and quantify structural changes in three civil engineering structures. The methodologies are evaluated based on computational costs and their abilities to identify structural anomalies accurately. Signals from the structures, collected by accelerometers, feed the autoencoders for unsupervised classification. The latent layer values of the autoencoders are used as parameters in Hotelling’s T2, and results are compared between classes to assess structural changes. Average execution times of each model were calculated for computational efficiency. Despite variations, computational cost did not hinder any methodology. The study demonstrates that the best fitting model, VAE-T2, outperforms its counterparts in identifying and quantifying structural changes. While the AE, SAE, and CAE models showed limitations in quantifying changes, they remain relevant for detecting anomalies. Continuous application and development of these techniques contribute to SHM advancements, enabling the increased safety, cost-effectiveness, and long-term durability of civil engineering structures.

This paper investigates the potential of utilising multilevel quasi-interpolation techniques on Chebyshev sparse grids for complex numerical computations. The paper starts by laying down the motivations for choosing Chebyshev sparse grids and quasi-interpolation methods with Gaussian kernels. It delves into the practical aspects of implementing these techniques. Various numerical experiments are performed to evaluate the efficiency and limitations of the multilevel quasi-sparse interpolation methods with dimensions two dimension and three dimension. The work ultimately aims to provide a comprehensive understanding of the computational efficiency and accuracy achievable through this approach, comparing its performance with traditional methods.

Purposes In the research of depression classification and diagnosis, feature selection plays a crucial role. Methods To address the issues of missing group effect information in existing hypergraph regularized feature selection methods, the group lasso-based hypergraph regularized feature selection approach is proposed. Specifically, the functional magnetic resonance imaging (fMRI) dataset is preprocessed first for depression. Second, on the basis of the preprocessed fMRI data, five brain network models under different scales are constructed and the topological attributes are calculated to extract features. After feature extracting, the group lasso method is introduced to build hypergraph and the hypergraph regularized feature selection method is employed to select features. At last, classification model is constructed by using support vector machine (SVM) and its performance is evaluated. Additionally, the effectiveness of the proposed method is validated on UCI datasets. Findings The demonstrate that the proposed method outperforms traditional feature selection methods across five different node templates. Moreover, for similar numbers of nodes in different templates, superior classification diagnostic performance is achieved.

The Dirichlet‐multinomial (DM) distribution plays a fundamental role in modern statistical methodology development and application. Recently, the DM distribution and its variants have been used extensively to model multivariate count data generated by high‐throughput sequencing technology in omics research due to its ability to accommodate the compositional structure of the data as well as overdispersion. A major limitation of the DM distribution is that it is unable to handle excess zeros typically found in practice which may bias inference. To fill this gap, we propose a novel Bayesian zero‐inflated DM model for multivariate compositional count data with excess zeros. We then extend our approach to regression settings and embed sparsity‐inducing priors to perform variable selection for high‐dimensional covariate spaces. Throughout, modeling decisions are made to boost scalability without sacrificing interpretability or imposing limiting assumptions. Extensive simulations and an application to a human gut microbiome dataset are presented to compare the performance of the proposed method to existing approaches. We provide an accompanying R package with a user‐friendly vignette to apply our method to other datasets. [ABSTRACT FROM AUTHOR]

In order to realize the automatic characterization and analysis of fabric texture images, a fabric defect detection method was proposed based on local texture image characterization. According to different applications of texture characterization and defect detection, the sparse cardinality and the sub window size were optimized. For the fabric texture without defects, overlapping sub windows can extract more texture elements to reconstruct the image. For defective fabrics, non overlapping sub window division can reduce the number of samples and the sparse cardinality, and improve the detection efficiency. The sub window size of 8 pixels and 8 pixels was selected for fabric texture characterization, while the sub window size of fabric defect detection was 32 pixels and 32 pixels. Experimental results show that overlapping sub windows can extract more texture elements, suppress Gaussian noise to a certain extent, so as to obtain high-quality reconstructed images. The defect detection rate of 58 sample images is 96.6%, which verifies that the algorithm has good adaptability to different fabric textures and defect types. [ABSTRACT FROM AUTHOR]

Copyright of Space-Integrated-Ground Information Networks is the property of Beijing Xintong Media Co., Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The classification problem is a key area of research in machine learning. The Least Squares Support Vector Machine (LSSVM) is an important classifier that is commonly used to solve classification problems. Its widespread use stems from its replacement of the inequality constraint in the Support Vector Machine (SVM) with the equality constraint, which transforms the convex quadratic programming (QP) problem of SVM into the solution of linear equations. However, when dealing with multi-class classification problems, LSSVM faces the challenges of lack of sparsity and sample noises, which can negatively impact its performance. Based on the modeling characteristics and data distribution of the multi-class LSSVM model, this paper proposes two improvements and establishes an improved fuzzy sparse multi-class least squares support vector machine (IF-S-M-LSSVM). The first improvement adopts a non-iterative sparse algorithm, which can delete training sample points to different degrees by adjusting the sparse ratio. The second improvement addresses the impact of sample noise on determining the optimal hyperplane by adding a fuzzy membership degree based on sample density. The advantages of the new model, in terms of training speed and classification accuracy, are verified through UCI machine learning standard data set experiments. Finally, the statistical significance of the IF-S-M-LSSVM model is tested using the Friedman and Bonferroni-Dunn tests. [ABSTRACT FROM AUTHOR]

With the continuous development of deep learning, long sequence time-series forecasting (LSTF) has attracted more and more attention in power consumption prediction, traffic prediction and stock prediction. In recent studies, various improved models of Transformer are favored. While these models have made breakthroughs in reducing the time and space complexity of Transformer, there are still some problems, such as the predictive power of the improved model being slightly lower than that of Transformer. And these models ignore the importance of special values in the time series. To solve these problems, we designed a more concise network named Resformer, which has four significant characteristics: (1) The fully sparse self-attention mechanism achieves O ⁢ (퐿푙표푔퐿) time complexity. (2) The AMS module is used to process the special values of time series and has comparable performance on sequences dependency alignment. (3) Using quadratic linear transformation, a simple LT module is designed to replace the self-attention mechanism. It effectively reduces redundant information. (4) The DistPooling method based on data distribution is proposed to suppress redundant information and noise. A large number of experiments on real data sets show that the Resformer method is superior to the existing improved model and standard Transformer method. [ABSTRACT FROM AUTHOR]

The main aim of this paper is to develop a nonconvex optimization model for third‐order tensor completion under wavelet transform. On the one hand, through wavelet transform of frontal slices, we divide a large tensor data into a main part tensor and three detail part tensors, and the elements of these four tensors are about a quarter of the original tensors. Solving these four small tensors can not only improve the operation efficiency, but also better restore the original tensor data. On the other hand, by using concave correction term, we are able to correct for low rank of tubal nuclear norm (TNN) data fidelity term and sparsity of l1$$ {l}_1 $$‐norm data fidelity term. We prove that the proposed algorithm can converge to some critical point. Experimental results on image, magnetic resonance imaging and video inpainting tasks clearly demonstrate the superior performance and efficiency of our developed method over state‐of‐the‐arts including the TNN and other methods. [ABSTRACT FROM AUTHOR]

In 5G communications, the acquisition of accurate channel state information (CSI) is of great importance to the hybrid beamforming of millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) system. In classical mmWave MIMO channel estimation methods, the exploitation of inherent sparse or low-rank structures has demonstrated to improve the performance. However, most high-accurate CSI estimators incur a high computational complexity and require the prior channel information, which hence present the major challenges in the practical deployment. In this work, we leverage machine learning to design the low-complexity and high-performance channel estimator. To be specific, we first formulate the CSI estimation, in the case of sparse structure, as one classical least absolute shrinkage and selection operator problem. In order to reduce the time complexity of existing compressed sensing (CS) methods, we then approximate the original optimization problem to another one, by imposing the other low-rank constraint that was barely considered by CS. We thus solve this new approximated problem and attain the near-optimal solution of the original problem. One new method excludes any prior channel information, and greatly improves the estimation performance, which only incurs a low time complexity. Simulation results demonstrate the superiority of our proposed method both in the estimation accuracy and time complexity. [ABSTRACT FROM AUTHOR]

Deep learning methods hold strong promise for identifying biomarkers for clinical application. However, current approaches for psychiatric classification or prediction do not allow direct interpretation of original features. In the present study, we introduce a sparse deep neural network (DNN) approach to identify sparse and interpretable features for schizophrenia (SZ) case-control classification. An L0 -norm regularization is implemented on the input layer of the network for sparse feature selection, which can later be interpreted based on importance weights. We applied the proposed approach on a large multi-study cohort with gray matter volume (GMV) and single nucleotide polymorphism (SNP) data for SZ classification. A total of 634 individuals served as training samples, and the classification model was evaluated for generalizability on three independent datasets of different scanning protocols (N = 394, 255, and 160, respectively). We examined the classification power of pure GMV features, as well as combined GMV and SNP features. Empirical experiments demonstrated that sparse DNN slightly outperformed independent component analysis + support vector machine (ICA + SVM) framework, and more effectively fused GMV and SNP features for SZ discrimination, with an average error rate of 28.98% on external data. The importance weights suggested that the DNN model prioritized to select frontal and superior temporal gyrus for SZ classification with high sparsity, with parietal regions further included with lower sparsity, echoing previous literature. The results validate the application of the proposed approach to SZ classification, and promise extended utility on other data modalities and traits which ultimately may result in clinically useful tools.

Efficiently removing clouds from remote sensing imagery presents a significant challenge, yet it is crucial for a variety of applications. This paper introduces a novel sparse function, named the tri-fiber-wise sparse function, meticulously engineered for the targeted tasks of cloud detection and removal. This function is adept at capturing cloud characteristics across three dimensions, leveraging the sparsity of mode-1, -2, and -3 fibers simultaneously to achieve precise cloud detection. By incorporating the concept of tensor multi-rank, which describes the global correlation, we have developed a tri-fiber-wise sparse-based model that excels in both detecting and eliminating clouds from images. Furthermore, to ensure that the cloud-free information accurately matches the corresponding areas in the observed data, we have enhanced our model with an extended box-constraint strategy. The experiments showcase the notable success of the proposed method in cloud removal. This highlights its potential and utility in enhancing the accuracy of remote sensing imagery.

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Among those who posit properties, liberals (mostly nominalists) admit abundant, ontologically free properties, which particulars possess whenever they satisfy the same predicate and belong to the same class, however artificial. I call them "L-properties" (for "Liberal"). Some liberals also admit that some few L-properties are natural, while most of them are artificial (the same applies to the corresponding classes). Others (mostly but not only realists) commit to a more discriminating use of the category: properties are sparse, they make for the objective similarities among particulars, and more importantly, they allow to analyze "objective similarity" and "class naturalness" in terms of property-possessing and sharing. They give particulars their "true nature", and I call them "A-properties" (for "Analytical"). This article provides a defense of this second type of properties, in the particular case of sortals. To that end, a new fact is put forward: that sorts are classes which are natural not as a primitive feature, but in virtue of what the particulars which belong to them are, and which makes them naturally belong to them. I then argue that this fact entails the existence of the desired type of properties, "sortal A-properties", although leaving open the question of how they should be construed. [ABSTRACT FROM AUTHOR]

Background: Lung cancer images require large memory storage and transmission bandwidth for sending the data. Compressive sensing (CS), as a method with a statistical approach in signal sampling, provides different output patterns based on information sources. Thus, it can be considered that CS can be used for feature extraction of compressed information. Methods: In this study, we proposed a novel texture extraction-based CS for lung cancer classification. We classify three types of lung cancer, including adenocarcinoma (ACA), squamous cell carcinoma (SCC), and benign lung cancer (N). The classification is carried out based on texture extraction, which is processed in 2 stages, the first stage to detect N and the second to detect ACA and SCC. Results: The simulation results show that two-stage texture extraction can improve accuracy by an average of 84%. The proposed system is expected to be decision support in assisting clinical diagnosis. In terms of technical storage, this system can save memory resources. Conclusions: The proposed two-step texture extraction system combined with CS and K- Nearest Neighbor has succeeded in classifying lung cancer with high accuracy; the system can also save memory storage. It is necessary to examine the complexity of the proposed method so that it can be analyzed further.

Background: With the rapid development of high-throughput sequencing technology and the explosive growth of genomic data, storing, transmitting and processing massive amounts of data has become a new challenge. How to achieve fast lossless compression and decompression according to the characteristics of the data to speed up data transmission and processing requires research on relevant compression algorithms. Methods: In this paper, a compression algorithm for sparse asymmetric gene mutations (CA_SAGM) based on the characteristics of sparse genomic mutation data was proposed. The data was first sorted on a row-first basis so that neighboring non-zero elements were as close as possible to each other. The data were then renumbered using the reverse Cuthill-Mckee sorting technique. Finally the data were compressed into sparse row format (CSR) and stored. We had analyzed and compared the results of the CA_SAGM, coordinate format (COO) and compressed sparse column format (CSC) algorithms for sparse asymmetric genomic data. Nine types of single-nucleotide variation (SNV) data and six types of copy number variation (CNV) data from the TCGA database were used as the subjects of this study. Compression and decompression time, compression and decompression rate, compression memory and compression ratio were used as evaluation metrics. The correlation between each metric and the basic characteristics of the original data was further investigated. Results: The experimental results showed that the COO method had the shortest compression time, the fastest compression rate and the largest compression ratio, and had the best compression performance. CSC compression performance was the worst, and CA_SAGM compression performance was between the two. When decompressing the data, CA_SAGM performed the best, with the shortest decompression time and the fastest decompression rate. COO decompression performance was the worst. With increasing sparsity, the COO, CSC and CA_SAGM algorithms all exhibited longer compression and decompression times, lower compression and decompression rates, larger compression memory and lower compression ratios. When the sparsity was large, the compression memory and compression ratio of the three algorithms showed no difference characteristics, but the rest of the indexes were still different. Conclusion: CA_SAGM was an efficient compression algorithm that combines compression and decompression performance for sparse genomic mutation data. [ABSTRACT FROM AUTHOR]

Factor model has the capacity of reducing redundant information in real data analysis. Note that sparse principal component (SPC) method is developed to obtain sparse solutions from the model, online principal component (OPC) method is used to handle with online dimension reduction problem. It is worth considering how to obtain a sparse solution with online learning. In this paper we propose a novel sparse online principal component (SOPC) method for sparse parameter estimation in factor model, where we combine the advantages of the SPC and OPC methods in estimating the loading matrix and the idiosyncratic variance matrix. By integrating sparse modelling with online update, the SOPC is capable of finding the sparse solution through iterative online updating, leading to a consistent and easily interpretable solution. Stability and sensitivity of the SOPC are assessed through a simulation study. The method is then applied to analyze two real data sets concerning drug efficacy and human activity recognition. [ABSTRACT FROM AUTHOR]

To pursue the outperformance of deep nets in learning, we construct a deep net with three hidden layers and prove that, implementing the empirical risk minimization (ERM) on this deep net, the estimator can theoretically realize the optimal learning rates without the classical saturation problem. In other words, deepening the networks with only three hidden layers can overcome the saturation and not degrade the optimal learning rates. The obtained results underlie the success of deep nets and provide a theoretical guidance for deep learning. [ABSTRACT FROM AUTHOR]

Introduction: The medical information contained in magnetic resonance imaging (MRI) and positron emission tomography (PET) has driven the development of intelligent diagnosis of Alzheimer’s disease (AD) and multimodal medical imaging. To solve the problems of severe energy loss, low contrast of fused images and spatial inconsistency in the traditional multimodal medical image fusion methods based on sparse representation. A multimodal fusion algorithm for Alzheimer’ s disease based on the discrete cosine transform (DCT) convolutional sparse representation is proposed. Methods: The algorithm first performs a multi-scale DCT decomposition of the source medical images and uses the sub-images of different scales as training images, respectively. Different sparse coefficients are obtained by optimally solving the sub-dictionaries at different scales using alternating directional multiplication method (ADMM). Secondly, the coefficients of high-frequency and low-frequency subimages are inverse DCTed using an improved L1 parametric rule combined with improved spatial frequency novel sum-modified SF (NMSF) to obtain the final fused images. Results and discussion: Through extensive experimental results, we show that our proposed method has good performance in contrast enhancement, texture and contour information retention. [ABSTRACT FROM AUTHOR]