Low-rank Feature Research Articles

Increased Vitamin Content or an Imbalance in Methyl Nutrients in the Gestational Diet Shifts the Gut Microbiota of the Offspring in a Sex-Dependent Manner

ObjectivesExcess gestational folic acid and insufficient choline intakes as observed in the North American populations may increase the risk of obesity in offspring. It is well-established that adverse health outcomes may arise due to shifts in the gut microbial communities, but whether high vitamin intakes or an imbalance between methyl nutrients contributes to gut microbiota alterations is unclear. The objective of this research was to determine the gut microbiota composition of male and female offspring in relation to the vitamin composition of the gestational diet. MethodsPregnant Wistar rats (n = 10/group) were fed the AIN-93G diet with either the recommended vitamin (RV), high multivitamin (HV), high folic acid (HFol) or high folic acid without choline (HFol-C) content. Male and female offspring were weaned to a high-fat control diet for 12 weeks. Fecal samples were collected from the colon upon termination for gut microbiota profiling by 16S rRNA sequencing and data analyses in QIIME2. ResultsThe overall gut microbial communities as assessed by unweighted UniFrac distances differed among the gestational diet groups for male (PERMANOVA P = 0.04) and female (PERMANOVA P = 0.05) offspring. The covariates gestational diet and sex predicted the gut microbiota differences in the offspring (Q2 = 0.07 in Songbird) whereas diet alone resulted in overfitting of the multinomial regression model (Q2 < 0). High ranked features from the natural log-ratios of microbial abundance were Shigella, Clostridiales, Clostridiaceae for HV, and Odoribacter, Akkermansia muciniphila, Blautia for both HFol and HFol-C compared to RV. Low ranked features were Odoribacter for HV, Clostridiaceae and Clostridiales for HFol, and Bifidobacterium, Allobaculum, Lactobacillus vaginalis for HFol-C compared to RV. In male offspring, Lactobacillus vaginalis, Sutterella and Clostridiales were high ranked and Odoribacter was low ranked compared to female offspring. These differentially abundant microbes may be important contributors to obesity across diet and sex. ConclusionsIncreased vitamin content or an imbalance between folic acid and choline in the gestational diet leads to a shift in the gut microbiota composition in the offspring toward obesity. These effects differed by sex. Funding SourcesUtah Agricultural Experiment Station and USU Research Catalyst. UNM supported by USU URCO.

Low-rank representation with adaptive dictionary learning for subspace clustering

High-dimensional data are often treated as collections of data samples approximately drawn from a union of multiple low-dimensional subspaces. Subspace clustering, where high-dimensional data samples are divided into low-dimensional subspace clusters, provides valuable insight into the underlying structures of high-dimensional data. The key challenge in subspace clustering is how to effectively measure the similarity among data samples. This paper presents an adaptive low-rank representation (ALRR) method for subspace clustering. An adaptive dictionary learning strategy that employs an orthonormality constraint is integrated into the low-rank representation (LRR) model. The dictionary, adaptively learned from the original data, makes the ALRR model robust to noise. The projection matrix and low-rank features are obtained simultaneously using an alternative optimization method. The convergence of ALRR is theoretically guaranteed under certain conditions, where ALRR requires at most three iterations for optimization. Consequently, it effectively obtains a convergence rate for ALRR that is better than those of several existing LRR algorithms. The experimental results on benchmark datasets show that the proposed method significantly outperforms several state-of-the-art subspace clustering methods, which indicates the effectiveness of ALRR for subspace clustering.

RIFS2D: A two-dimensional version of a randomly restarted incremental feature selection algorithm with an application for detecting low-ranked biomarkers

The era of big data introduces both opportunities and challenges for biomedical researchers. One of the inherent difficulties in the biomedical research field is to recruit large cohorts of samples, while high-throughput biotechnologies may produce thousands or even millions of features for each sample. Researchers tend to evaluate the individual correlation of each feature with the class label and use the incremental feature selection (IFS) strategy to select the top-ranked features with the best prediction performance. Recent experimental data showed that a subset of continuously ranked features randomly restarted from a low-ranked feature (an RIFS block) may outperform the subset of top-ranked features. This study proposed a feature selection Algorithm RIFS2D by integrating multiple RIFS blocks. A comprehensive comparative experiment was conducted with the IFS, RIFS and existing feature selection algorithms and demonstrated that a subset of low-ranked features may also achieve promising prediction performance. This study suggested that a prediction model with promising performance may be trained by low-ranked features, even when top-ranked features did not achieve satisfying prediction performance. Further comparative experiments were conducted between RIFS2D and t-tests for the detection of early-stage breast cancer. The data showed that the RIFS2D-recommended features achieved better prediction accuracy and were targeted by more drugs than the t-test top-ranked features.

Semisupervised Classification Based on Tensor Convolutional Neural Network for Hyperspectral Images

Deep neural network has been extensively applied to hyperspectral image (HSI) classification and shown promising performance recently. However, those popular deep learning models scarcely consider low-rank features of high-dimensional HSIs lying in intrinsic data subspaces. Besides, their results largely rely on numerous labeled samples, which are expensive and time-consuming. To address these issues, we propose a Three Dimensional Tensor Convolutional Neural Network (3DTCNN) for spectral-spatial low-rank feature representation and a Semisupervised Classifier based on 3DTCNN (S-TCNN) for classification with limited labeled samples. 3DTCNN integrates tensor decomposition with convolutional neural network to extract discriminative spectral-spatial low-rank features. It also combines multiple supervised discriminators and unsupervised clustering modules to exploit labeled and unlabeled samples. Experimental results show that S-TCNN outperforms several classic supervised classifiers and state-of-the-art semisupervised classifiers on one benchmark hyperspectral data set, Indian Pines. When there are only 3 samples with prior labels per class, the proposed model achieves 98.2%, in terms of overall accuracy. Therefore, the design of 3DTCNN is proved feasible while the efficacy and efficiency of S-TCNN are validated.

Collaborative Matrix Factorization with Soft Regularization for Drug-Target Interaction Prediction

Identifying the potential drug-target interactions (DTI) is critical in drug discovery. The drug-target interaction prediction methods based on collaborative filtering have demonstrated attractive prediction performance. However, many corresponding models cannot accurately express the relationship between similarity features and DTI features. In order to rationally represent the correlation, we propose a novel matrix factorization method, so-called collaborative matrix factorization with soft regularization (SRCMF). SRCMF improves the prediction performance by combining the drug and the target similarity information with matrix factorization. In contrast to general collaborative matrix factorization, the fundamental idea of SRCMF is to make the similarity features and the potential features of DTI approximate, not identical. Specifically, SRCMF obtains low-rank feature representations of drug similarity and target similarity, and then uses a soft regularization term to constrain the approximation between drug (target) similarity features and drug (target) potential features of DTI. To comprehensively evaluate the prediction performance of SRCMF, we conduct cross-validation experiments under three different settings. In terms of the area under the precision-recall curve (AUPR), SRCMF achieves better prediction results than six state-of-the-art methods. Besides, under different noise levels of similarity data, the prediction performance of SRCMF is much better than that of collaborative matrix factorization. In conclusion, SRCMF is robust leading to performance improvement in drug-target interaction prediction.

PolSAR Image Classification Based on Robust Low-Rank Feature Extraction and Markov Random Field

Polarimetric synthetic aperture radar (PolSAR) image classification has been investigated vigorously in various remote sensing applications. However, it is still a challenging task nowadays. One significant barrier lies in the speckle effect embedded in the PolSAR imaging process, which greatly degrades the quality of the images and further complicates the classification. To this end, we present a novel PolSAR image classification method that removes speckle noise via low-rank (LR) feature extraction and enforces smoothness priors via the Markov random field (MRF). Especially, we employ the mixture of Gaussian-based robust LR matrix factorization to simultaneously extract discriminative features and remove complex noises. Then, a classification map is obtained by applying a convolutional neural network with data augmentation on the extracted features, where local consistency is implicitly involved, and the insufficient label issue is alleviated. Finally, we refine the classification map by MRF to enforce contextual smoothness. We conduct experiments on two benchmark PolSAR data sets. Experimental results indicate that the proposed method achieves promising classification performance and preferable spatial consistency.

Multi-scale channel network based on filer pruning for image super-resolution

Image super-resolution is an important image restoration technology, which is widely used in optical system, optical communication and other fields. Deep convolutional neural networks (CNNs) have made a great breakthrough in accuracy of single image super-resolution. Nevertheless, as the depth and width of the networks increase, CNNs methods face the challenge of memory consumption and computational complexity. To solve this problem, in this paper, we introduce a multi-scale channel network (MSCN) based on filter pruning to reconstruct the high-resolution (HR) image from the original low-resolution (LR) image. Firstly, we propose a multi-scale channel block (MSCB) to obtain the fine features based on the convolution kernels with different receptive fields. Channel split operator is used in MSCB to reduce the model parameters. The wider MSCN with more filters is trained to achieve a better performance. Then, we calculate the rank of feature maps and prune filters with low-rank feature information. Finally, the compressed multi-scale channel network with fewer parameters can achieve similar accuracies. The experimental results validate the effectiveness of our approaches.

Millimeter Wave Channel Estimation using Alternating Direction Method of Multipliers

With the explosive growth in demand for mobile data traffic, the contradiction between capacity requirements and spectrum scarcity becomes more and more prominent. The bandwidth is becoming a key issue in 5G mobile networks. However, with the huge bandwidth from 30 GHz to 300 GHz, mm Wave communications considered an important part of the 5G mobile network providing multi communication services, where channel state information considers a challenging task for millimeter wave MIMO systems due to the huge number of antennas. Therefore, this paper discusses the channel and signal models of the mm Wave, with a novel formulation for mm Wave channel estimation inclusive low rank features, that we improved using a developed theory of matrix completion with Alternating Direction Method.

Data recoverability and estimation for perception layer in semantic web of things.

Internet of Things (IoT) is the growing invention in the current development of different domains like industries, e-health, and education, etc. Semantic web of things (SWoT) is an extension of IoT that enhance the communication by behaving intelligently. SWoT comprises 7 layered architecture. The perception layer is an important layer for collecting data from devices and to communicate with its associated layer. The data loss at the perception layer is very common due to inadequate resources, unpredictable link, noise, collision, and unexpected damage. To address this problem, we propose a method based on Compressive Sensing which recovers and estimates sensory data from a low-rank structure. The contribution of this paper is three folds. Firstly, we determine the problem of data acquisition and data loss at semantic sensory nodes in SWoT. Secondly, we introduce a compressive sensing based framework for SWoT that recovers the data accurately using low-rank features. Thirdly, the data estimation method is utilized to reduce the volume of the data. Proposed Compressive Sensing based Data Recoverability and Estimation (CS-RE) method is evaluated and compared with the existing reconstruction methods. The simulation results on real sensory datasets depict that the proposed method significantly outperforms existing methods in terms of error ratio and data recoverability accuracy.

Airborne Radar Super-Resolution Imaging Based on Fast Total Variation Method

Total variation (TV) is an effective super-resolution method to improve the azimuth resolution and preserve the contour information of the target in airborne radar imaging. However, the computational complexity is very high because of the matrix inversion, reaching O(N3). In this paper, a Gohberg–Semencul (GS) representation based fast TV (GSFTV) method is proposed to make up for the shortcoming. The proposed GSFTV method fist utilizes a one-dimensional TV norm as the regular term under regularization framework, which is conducive to achieve super-resolution while preserving the target contour. Then, aiming at the very high computational complexity caused by matrix inversion when minimizing the TV regularization problem, we use the low displacement rank feature of Toeplitz matrix to achieve fast inversion through GS representation. This reduces the computational complexity from O(N3) to O(N2), benefiting efficiency improvement for airborne radar imaging. Finally, the simulation and real data processing results demonstrate that the proposed GSFTV method can simultaneously improve the resolution and preserve the target contour. Moreover, the very high computational efficiency of the proposed GSFTV method is tested by hardware platform.

A Superfast Super-Resolution Method for Radar Forward-Looking Imaging

The super-resolution method has been widely used for improving azimuth resolution for radar forward-looking imaging. Typically, it can be achieved by solving an undifferentiable regularization problem. The split Bregman algorithm (SBA) is a great tool for solving this undifferentiable problem. However, its real-time imaging ability is limited to matrix inversion and iterations. Although previous studies have used the special structure of the coefficient matrix to reduce the computational complexity of each iteration, the real-time performance is still limited due to the need for hundreds of iterations. In this paper, a superfast SBA (SFSBA) is proposed to overcome this shortcoming. Firstly, the super-resolution problem is transmitted into an regularization problem in the framework of regularization. Then, the proposed SFSBA is used to solve the nondifferentiable regularization problem. Different from the traditional SBA, the proposed SFSBA utilizes the low displacement rank features of Toplitz matrix, along with the Gohberg-Semencul (GS) representation to realize fast inversion of the coefficient matrix, reducing the computational complexity of each iteration from to . It uses a two-order vector extrapolation strategy to reduce the number of iterations. The convergence speed is increased by about 8 times. Finally, the simulation and real data processing results demonstrate that the proposed SFSBA can effectively improve the azimuth resolution of radar forward-looking imaging, and its performance is only slightly lower compared to traditional SBA. The hardware test shows that the computational efficiency of the proposed SFSBA is much higher than that of other traditional super-resolution methods, which would meet the real-time requirements in practice.

LR-Net: Low-Rank Spatial-Spectral Network for Hyperspectral Image Denoising.

Due to the physical limitations of the imaging devices, hyperspectral images (HSIs) are commonly distorted by a mixture of Gaussian noise, impulse noise, stripes, and dead lines, leading to the decline in the performance of unmixing, classification, and other subsequent applications. In this paper, we propose a novel end-to-end low-rank spatial-spectral network (LR-Net) for the removal of the hybrid noise in HSIs. By integrating the low-rank physical property into a deep convolutional neural network (DCNN), the proposed LR-Net simultaneously enjoys the strong feature representation ability from DCNN and the implicit physical constraint of clean HSIs. Firstly, spatial-spectral atrous blocks (SSABs) are built to exploit spatial-spectral features of HSIs. Secondly, these spatial-spectral features are forwarded to a multi-atrous block (MAB) to aggregate the context in different receptive fields. Thirdly, the contextual features and spatial-spectral features from different levels are concatenated before being fed into a plug-and-play low-rank module (LRM) for feature reconstruction. With the help of the LRM, the workflow of low-rank matrix reconstruction can be streamlined in a differentiable manner. Finally, the low-rank features are utilized to capture the latent semantic relationships of the HSIs to recover clean HSIs. Extensive experiments on both simulated and real-world datasets were conducted. The experimental results show that the LR-Net outperforms other state-of-the-art denoising methods in terms of evaluation metrics and visual assessments. Particularly, through the collaborative integration of DCNNs and the low-rank property, the LR-Net shows strong stability and capacity for generalization.

Efficient update of multilinear singular value decomposition in background subtraction applications

Subtraction techniques are used to distinguish moving objects or foregrounds that are being tracked from a static background. To prevent possible local misclassifications, the subspace spanned by low-rank features computed from a multilinear singular value decomposition (MLSVD), can be used to filter out noises or gradual changes from the background. However, as it is prohibitively expensive to compute a new MLSVD from scratch at every iteration, we propose an adaptive and efficient method for updating an MLSVD by reusing previous decompositions while tracking more accurate decomposition errors. The experimental results reveal that the proposed MLSVD update algorithm exhibits a faster execution speed and better accuracy than other MLSVD update algorithms used in background subtraction applications.

The dynamic change of pore structure for low-rank coal under refined upgrading pretreatment temperatures

Pore structure characteristics are significant factor in the evaluation of the physical characteristics of low-rank coal. In this study, three low-rank coal samples were collected from the Xishanyao Formation, Santanghu Basin, and low-temperature liquid-nitrogen adsorption (LP-N2A) measurements were taken under various pretreatment temperatures. Owing to the continuous loss of water and volatile matter in low-rank coal, the total pore volume assumes a three-step profile with knee temperatures of 150 °C and 240 °C. However, the ash in the coal can protect the coal skeleton. Pore collapse mainly occurs for mesopores with aperture smaller than 20 nm. Mesopores with apertures smaller than 5 nm exhibit a continuous decrease in pore volume, whereas the pore volume of mesopores with apertures ranging from 5 to 10 nm increases at lower pretreatment temperatures (<150 °C) followed by a faint decrease. As for mesopores with apertures larger than 10 nm, the pore volume increases significantly when the pretreatment temperature reaches 300 °C. The pore structure of low-rank coal features a significant heating effect, the pretreatment temperature should not exceed 150 °C when the LP-N2A is used to evaluate the pore structure of low-rank coal to effectively evaluate the reservoir characteristics of low-rank coal.

Unsteady reduced-order model of flow over cylinders based on convolutional and deconvolutional neural network structure

In this paper, we propose a neural network based reduced-order model for predicting the unsteady flow field over single/multiple cylinders. The neural network model constructs a mapping function between the temporal evolution of the pressure signal on the cylinder surface and the surrounding velocity field, where Convolutional Neural Network (CNN) layers are used as the encoder and deconvolutional neural network layers are used as the decoder. Compared with the network model with a fully connected (FC) decoder, the model with the deconvolution connected (DC) decoder is good for capturing and reconstructing the spatial relationships of low-rank feature spaces, such as edge intersections, parallelism, and symmetry, while the fluid flow, which is described by Navier–Stokes equations containing convection and diffusion terms, displays outstanding features of locality. In this article, the performance of the network models with the FC decoder and the DC decoder is evaluated by studying the problem of flow over a single cylinder first, and then the complexity of the flow structure of the studied problems is enhanced by increasing the number of cylinders and the Reynolds number. The results indicate that both the CNN-FC decoder model and CNN-DC decoder model achieve fast and accurate prediction on the velocity field, and the CNN-DC decoder model gives more robust and precise performance for all studied problems.

Intelligent Detection for Key Performance Indicators in Industrial-Based Cyber-Physical Systems

Intelligent anomaly detection for key performance indicators (KPIs) is important for keeping services reliable in industrial-based cyber-physical systems (CPS). However, it is common in practice for various KPI sampling strategies to be utilized. We experimentally verify that anomaly detection is highly sensitive to irregular sampling, and accordingly go on to investigate low-cost anomaly detection for large-scale irregular KPIs. Irregular KPIs can be classified into four types: equal interval and unequal quantity (EIUQ) KPIs, unequal interval (UI) KPIs, unequal interval with equal duration (UIED) KPIs, and segmented irregular KPIs. In this article, we propose an anomaly detection framework based on these irregular types. Moreover, to handle the various lengths and phase shifts among EIUQ KPIs, we propose a normalized version of unequal cross-correlation, which slides the KPIs to enable finding the most similar position. To avoid high computational costs, we analyze the low-rank feature of KPIs data and propose a matrix factorization-based alignment algorithm for UIED KPIs; this algorithm treats UIED KPIs as an incomplete matrix and recovers the KPIs to align them before performing anomaly detection. Extensive simulations using three public datasets and two real-world datasets demonstrate that our algorithm can achieve a larger F1-score than Minkowski distance and less time than dynamic time warping distance.

Superresolution of Radar Forward-Looking Imaging Based on Accelerated TV-Sparse Method

Total variation-sparse (TV-sparse)-based multiconstraint devonvolution method has been used to realize superresolution imaging and preserve target contour information simultaneously of radar forward-looking imaging. However, due to the existence of matrix inversion, it suffers from high computational complexity, which restricts the ability of radar real-time imaging. In this article, an Gohberg-Semencul (GS) decomposition-based fast TV-sparse (FTV-sparse) method is proposed to reduce the computational complexity of TV-sparse method. The acceleration strategy utilizes the low displacement rank features of Toeplitz matrix, realizing fast matrix inversion by using a GS representation. It reduces the computational complexity of traditional TV-sparse method from O(N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> ) to O(N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ), benefiting for improvement of the computing efficiency. The simulation and experimental data processing results show that the proposed FTV-sparse method has almost no resolution loss compared with the traditional TV sparse method. Hardware test results show that the proposed FTV-sparse method significantly improves the computational efficiency of TVsparse method.

Small Target Detection for Infrared Image Based on Optimal Infrared Patch-Image Model by Solving Modified Adaptive RPCA Problem

Small target detection in infrared (IR) images has been widely applied for both military and civilian purposes. In this study, because IR images contain sparse and low-rank features in most scenarios, we propose an optimal IR patch-image (OIPI) model-based detection method to detect small targets in heavily cluttered IR images. First, the OIPI model was generated based on a conventional IR image model using a novel optimal patch size and sliding step adaptive selection algorithm. Secondly, the sparse and low-rank features of IR images were extracted and fused to generate an adaptive weighted parameter. Thirdly, the adaptive inexact augmented Lagrange multiplier (AIALM) algorithm was applied in the OIPI model to solve the robust principal component analysis (RPCA) optimization problem. Finally, an adaptive threshold method is proposed to segment and calibrate targets. Experimental results indicate that the proposed algorithm is capable of detecting small targets more stably and accurately, compared with state-of-the-art methods.

ADRML: anticancer drug response prediction using manifold learning

One of the prominent challenges in precision medicine is to select the most appropriate treatment strategy for each patient based on the personalized information. The availability of massive data about drugs and cell lines facilitates the possibility of proposing efficient computational models for predicting anticancer drug response. In this study, we propose ADRML, a model for Anticancer Drug Response Prediction using Manifold Learning to systematically integrate the cell line information with the drug information to make accurate predictions about drug therapeutic. The proposed model maps the drug response matrix into the lower-rank spaces that lead to obtaining new perspectives about cell lines and drugs. The drug response for a new cell line-drug pair is computed using the low-rank features. The evaluation of ADRML performance on various types of cell lines and drug information, in addition to the comparisons with previously proposed methods, shows that ADRML provides accurate and robust predictions. Further investigations about the association between drug response and pathway activity scores reveal that the predicted drug responses can shed light on the underlying drug mechanism. Also, the case studies suggest that the predictions of ADRML about novel cell line-drug pairs are validated by reliable pieces of evidence from the literature. Consequently, the evaluations verify that ADRML can be used in accurately predicting and imputing the anticancer drug response.

Fast Split Bregman Based Deconvolution Algorithm for Airborne Radar Imaging

Deconvolution methods can be used to improve the azimuth resolution in airborne radar imaging. Due to the sparsity of targets in airborne radar imaging, an L 1 regularization problem usually needs to be solved. Recently, the Split Bregman algorithm (SBA) has been widely used to solve L 1 regularization problems. However, due to the high computational complexity of matrix inversion, the efficiency of the traditional SBA is low, which seriously restricts its real-time performance in airborne radar imaging. To overcome this disadvantage, a fast split Bregman algorithm (FSBA) is proposed in this paper to achieve real-time imaging with an airborne radar. Firstly, under the regularization framework, the problem of azimuth resolution improvement can be converted into an L 1 regularization problem. Then, the L 1 regularization problem can be solved with the proposed FSBA. By utilizing the low displacement rank features of Toeplitz matrix, the proposed FSBA is able to realize fast matrix inversion by using a Gohberg–Semencul (GS) representation. Through simulated and real data processing experiments, we prove that the proposed FSBA significantly improves the resolution, compared with the Wiener filtering (WF), truncated singular value decomposition (TSVD), Tikhonov regularization (REGU), Richardson–Lucy (RL), iterative adaptive approach (IAA) algorithms. The computational advantage of FSBA increases with the increase of echo dimension. Its computational efficiency is 51 times and 77 times of the traditional SBA, respectively, for echoes with dimensions of 218 × 400 and 400 × 400 , optimizing both the image quality and computing time. In addition, for a specific hardware platform, the proposed FSBA can process echo of greater dimensions than traditional SBA. Furthermore, the proposed FSBA causes little performance degradation, when compared with the traditional SBA.