Screening Rules Research Articles

Safe screening rules for multi-view support vector machines

Multi-view learning aims to make use of the advantages of different views to complement each other and fully mines the potential information in the data. However, the complexity of multi-view learning algorithm is much higher than that of single view learning algorithm. Based on the optimality conditions of two classical multi-view models: SVM-2K and multi-view twin support vector machine (MvTwSVM), this paper analyzes the corresponding relationship between dual variables and samples, and derives their safe screening rules for the first time, termed as SSR-SVM-2K and SSR-MvTwSVM. It can assign or delete four groups of different dual variables in advance before solving the optimization problem, so as to greatly reduce the scale of the optimization problem and improve the solution speed. More importantly, the safe screening criterion is “safe”, that is, the solution of the reduced optimization problem is the same as that of the original problem before screening. In addition, we further give a sequence screening rule to speed up the parameter optimization process, and analyze its properties, including the similarities and differences of safe screening rules between multi-view SVMs and single-view SVMs, the computational complexity, and the relationship between the parameter interval and screening rate. Numerical experiments verify the effectiveness of the proposed methods.

Explicit molecule-based reaction network simplification: Theory and application on catalytic reforming

The complex molecular reaction network may contain excessive reaction details, resulting in the difficulty in solving the model. To reduce the size of a complex reaction network, this work proposed a reaction network simplification algorithm. The explicit molecules can be determined from the network according to the molecular-level characterization information. The remaining non-explicit substances were defined as intermediate and terminal molecules, respectively. A reaction screening rule was proposed to select key reactions and reduce the reaction network. This work took naphtha catalytic reforming as a studying case, and a reaction network containing excessive details was generated via reaction rules. After the reaction network simplification algorithm was employed, a simplified network was obtained. The results indicate that the product yield and temperature distribution calculated through the reduced network has an excellent agreement with the complete network, and the product composition also can be computed accurately. Simultaneously, the computational efficiency is enhanced.

Multi-task twin bounded support vector machine and its safe screening rule

Multi-task twin bounded support vector machine and its safe screening rule

Multi-parameter nuclear data adjustment for sodium fast reactor

This paper studies the application of nuclear data adjustment method in the physics calculation of sodium cooled fast reactor. Based on Bayesian theory, the multi-parameter nuclear data adjustment method is derived, and then the experimental screening rules are given. The verification based on 47 candidate experiments, including the critical measurement experiments, control rod worth measurement experiments and sodium void reactivity measurement experiments of ZPPR-9、ZPPR-10A、ZPPR-10B and ZPPR-10C zero-power experimental facilities, is then carried out. The numerical results show that the adjustment based only on keff has obvious limitations, it can only reduce the calculation uncertainty of keff, but the uncertainties of other physical parameters cannot be improved effectively. However, the multi-parameter nuclear data adjustment method effectively reduces the calculation uncertainties and errors of different physical parameters besides keff. In the meantime, the screening rules for experimental data are analyzed to avoid the improper adjustment of nuclear data.

Elastic net twin support vector machine and its safe screening rules

In this paper, we present a new classifier called elastic net twin support vector machine (ETSVM). It resolves two smaller-sized quadratic programming problems (QPPs) similarly to the twin support vector machine (TSVM). The key difference between them is that ETSVM uses elastic net penalty for slack variables, which enhances classification performance. The dual QPPs of the ETSVM do not involve matrix inversion, in contrast to conventional TSVM. As a result, ETSVM can avoid the ill-conditioning case. We theoretically discuss its properties, including exploring the violation tolerance upper bound for the two QPPs of ETSVM. In order to increase computing efficiency, we derive a sequence of safe screening rules using variational inequalities to quicken the ETSVM parameter tuning process (SSR-ETSVM). We compare the proposed algorithm with SVM, nonparallel hyperplane SVM, elastic net SVM, and elastic net nonparallel hyperplane SVM. Numerical experiments confirm the rationality and effectiveness of the proposed methods.

Label Pair of Instances-Based Safe Screening for Multilabel Rank Support Vector Machine

Rank support vector machine (RSVM) is widely used in multilabel classification problems. However, as the number of labels and instances soars, the training efficiency of the model will be greatly reduced. Unfortunately, few effective methods can solve this problem. In this article, we propose a safe screening rule (SSR) for RSVM to improve its training speed. This is the first attempt to construct SSR for multilabel learning problems. SSR for RSVM can screen and delete most of the instances based on their relevant–irrelevant label pairs, which is the biggest difference in the existing SSR. After this process, the scale of RSVM can be substantially reduced before solving it. The sequential version of SSR for RSVM is further introduced to accelerate the whole parameter tuning process. One important advantage of SSR is that it is safe, which means we can obtain the same optimal solution as the original problem by utilizing it. Extensive experiments with five benchmark datasets, three large-scale datasets, and one type 2 diabetes dataset show that our approach is efficient and safe.

Subspace screening rule for multi-label estimator with sparsity-inducing regularization

Multi-label learning has drawn wide attention for the last decade. To exploit the correlation between labels, a multi-label classifier based on the nuclear norm has been proposed recently, which joints Ranking support vector machine (RankSVM) and Binary Relevance (BR) with robust Low-rank learning (RBRL). Therefore, it has satisfactory classification outcomes in the application. Nonetheless, tackling the large-scale problem still remains a challenge for RBRL. Motivated by this, a Subspace Screening Rule (SSR) for RBRL is proposed to accelerate the solving process. Its primary strategy is to reduce the size of the matrix variable to be estimated, based on the fact that a low-rank matrix can be represented by a few subspaces. Specifically, at each iteration we delete a majority of subspaces with zero coefficients in the optimal solution by matrix decomposition and optimality condition. Then, we solve the small-scale reduced problem rather than the initial large-scale matrix problem. An excellent acceleration effect is obtained. To further improve the solving speed, Approximate Singular Value Decomposition (ASVD) and Accelerated Proximal Gradient (APG) are employed in the different stages. Extensive experiments on seven benchmark datasets as well as an artificial dataset demonstrate the efficiency of SSR.

RCFGL: Rapid Condition adaptive Fused Graphical Lasso and application to modeling brain region co-expression networks.

Inferring gene co-expression networks is a useful process for understanding gene regulation and pathway activity. The networks are usually undirected graphs where genes are represented as nodes and an edge represents a significant co-expression relationship. When expression data of multiple (p) genes in multiple (K) conditions (e.g., treatments, tissues, strains) are available, joint estimation of networks harnessing shared information across them can significantly increase the power of analysis. In addition, examining condition-specific patterns of co-expression can provide insights into the underlying cellular processes activated in a particular condition. Condition adaptive fused graphical lasso (CFGL) is an existing method that incorporates condition specificity in a fused graphical lasso (FGL) model for estimating multiple co-expression networks. However, with computational complexity of O(p2K log K), the current implementation of CFGL is prohibitively slow even for a moderate number of genes and can only be used for a maximum of three conditions. In this paper, we propose a faster alternative of CFGL named rapid condition adaptive fused graphical lasso (RCFGL). In RCFGL, we incorporate the condition specificity into another popular model for joint network estimation, known as fused multiple graphical lasso (FMGL). We use a more efficient algorithm in the iterative steps compared to CFGL, enabling faster computation with complexity of O(p2K) and making it easily generalizable for more than three conditions. We also present a novel screening rule to determine if the full network estimation problem can be broken down into estimation of smaller disjoint sub-networks, thereby reducing the complexity further. We demonstrate the computational advantage and superior performance of our method compared to two non-condition adaptive methods, FGL and FMGL, and one condition adaptive method, CFGL in both simulation study and real data analysis. We used RCFGL to jointly estimate the gene co-expression networks in different brain regions (conditions) using a cohort of heterogeneous stock rats. We also provide an accommodating C and Python based package that implements RCFGL.

Feature screening strategy for non-convex sparse logistic regression with log sum penalty

L1 logistic regression is an efficient classification algorithm. Leveraging on the convexity of the model and the sparsity of L1 norm, techniques named safe screening rules can help accelerate solvers for high-dimensional data. Recently, Log Sum Penalty (LSP) provides better theoretical guarantees in identifying relevant variables, and sparse logistic regression with LSP has been proposed. However, due to the non-convexity of this model, the training process is time-consuming. Furthermore, the existing safe screening rules cannot be directly applied to accelerate it. To deal with this issue, in this paper, based on the iterative majorization minimization (MM) principle, we construct a novel method that can effectively save training time. To do this, we first design a feature screening strategy for the inner solver and then build another rule to propagate screened features between the iterations of MM. After that, we introduce a modified feature screening strategy to further accelerate the computational speed, which can obtain a smaller safe region thanks to reconstructing the strong global concavity bound. Moreover, our rules can be applied to other non-convex cases. Experiments on nine benchmark datasets verify the effectiveness and security of our algorithm.

Screening based sparse classification method for the detection of corn leaf diseases

Detection of diseases in corn plant at early stage reduces the chance of productivity loss. Recognizing the different kind of disease in corn plant requires the identification of lesion part of the plant along with the extraction of proper features from the region of interest. Although several imaging tools and classifier were developed in the recent past towards identifying different kinds of disease in corn plant but most of the classification models do not meet the specific requirement due to improper selection of feature and classifier. In this paper, we proposed a two-step based sparse recognition model for the identification of corn disease. First step emphasizes with the construction of image dictionary by extracting multiple features from the lesion part of training image and followed it in the second phase, uses the screening and angle rule based sparse classifier to recognize the different corn diseases. Utilizing the data from the kaggle dataset and comparing with the other state of art, our method yielded an average classification accuracy of 88.55%. This makes the framework feasible and effective for the detection of corn plant disease.

Research on the Evolution of Journal Topic Mining Based on the BERT-LDA Model

Scientific papers are an important form for researchers to summarize and display their research results. Information mining and analysis of scientific papers can help to form a comprehensive understanding of the subject. Aiming at the ignorance of contextual semantic information in current topic mining and the uncertainty of screening rules in association evolution research, this paper proposes a topic mining evolution model based on the BERT-LDA model. First, the model combines the contextual semantic information learned by the BERT model with the word vectors of the LDA model to mine deep semantic topics. Then construct topic filtering rules to eliminate invalid associations between topics. Finally, the relationship between themes is analyzed through the theme evolution, and the complex relationship between the themes such as fusion, diffusion, emergence, and disappearance is displayed. The experimental results show that, compared with the traditional LDA model, the topic mining evolution model based on BERTLDA can accurately mine topics with deep semantics and effectively analyze the development trend of scientific and technological paper topics.

A two-stage gap safe screening rule for multi-label optimal margin distribution machine

Multi-label optimal margin distribution machine (mlODM) is an efficient algorithm for multi-label classification. Although it can achieve great generalization performance, it is inefficient for large-scale datasets due to the huge number of label pairs. Motivated by its sparse solution, in this paper, we propose a two-stage gap safe screening rule for accelerating mlODM, termed as TSSR. First, a sequential safe screening rule (SSSR) based on gap is designed to screen out part of redundant label pairs prior to training, which reduces the scale of mlODM. Compared with the previous DVI rule, our method ensures absolute safety without destroying efficiency. In the second stage, to further speed up the solving process, a dynamic safe screening rule (DSSR) is embedded into the solving algorithm DCDM when training the simplified mlODM. More importantly, the feasible solution generated in the first stage can promote the efficiency of DSSR. To the best of our knowledge, this is the first attempt to create a hybrid screening rule for multi-label model. Our TSSR can greatly reduce the cost and achieve exactly the same accuracy. Experimental results on seven multi-label benchmark datasets and two real-world learning problems including movie genres classification and hypoglycemic drugs prediction verify the superiority of the proposed methods.

Elastic Net Nonparallel Hyperplane Support Vector Machine and Its Geometrical Rationality.

Twin support vector machine (TWSVM), which constructs two nonparallel classifying hyperplanes, is widely applied to various fields. However, TWSVM solves two quadratic programming problems (QPPs) separately such that the final classifiers lack consistency and enough prediction accuracy. Moreover, by reason of only considering the 1-norm penalty for slack variables, TWSVM is not well defined in the geometrical view. In this article, we propose a novel elastic net nonparallel hyperplane support vector machine (ENNHSVM), which adopts elastic net penalty for slack variables and constructs two nonparallel separating hyperplanes simultaneously. We further discuss the properties of ENNHSVM theoretically and derive the violation tolerance upper bound to better demonstrate the relative violations of training samples in the same class. In particular, we design a safe screening rule for ENNHSVM to speed up the calculations. We finally compare the performance of ENNHSVM on both synthetic datasets and benchmark datasets with the Lagrangian SVM, the twin parametric-margin SVM, the elastic net SVM, the TWSVM, and the nonparallel hyperplane SVM.

Penalized logistic regression with prior information for microarray gene expression classification.

Cancer classification and gene selection are important applications in DNA microarray gene expression data analysis. Since DNA microarray data suffers from the high-dimensionality problem, automatic gene selection methods are used to enhance the classification performance of expert classifier systems. In this paper, a new penalized logistic regression method that performs simultaneous gene coefficient estimation and variable selection in DNA microarray data is discussed. The method employs prior information about the gene coefficients to improve the classification accuracy of the underlying model. The coordinate descent algorithm with screening rules is given to obtain the gene coefficient estimates of the proposed method efficiently. The performance of the method is examined on five high-dimensional cancer classification datasets using the area under the curve, the number of selected genes, misclassification rate and F-score measures. The real data analysis results indicate that the proposed method achieves a good cancer classification performance with a small misclassification rate, large area under the curve and F-score by trading off some sparsity level of the underlying model. Hence, the proposed method can be seen as a reliable penalized logistic regression method in the scope of high-dimensional cancer classification.

Novel Method for Comprehensive Annotation of Plant Glycosides Based on Untargeted LC-HRMS/MS Metabolomics.

Glycosides are a large family of secondary metabolites in plants, which play a critical role in plant growth and development. Due to the complexity and diversity in structures and the limited availability of authentic standards, comprehensive annotation of the glycosides remains a great challenge. In this study, using maize as an example, a deep annotation method of glycosides was proposed based on untargeted liquid chromatography-high-resolution tandem mass spectrometry metabolomics analysis. First, knowledge-based in silico aglycone and glycosyl/acyl-glycosyl libraries were built. A total of 1240 known and potential aglycones from databases and literature were recorded. Next, the MS parameters beneficial to aglycone ion-rich MS/MS were explored using 1782 high-resolution MS/MS spectra of glycosides from the MassBank of North America (MoNA) and confirmed by 52 authentic glycoside standards. Then, screening rules for aglycon ions in MS/MS were recommended. Glycoside candidates were further filtered by MS/MS-based chemical classification and MS/MS similarity of aglycon-glycoside pairs. Finally, the glycosylation sites of flavonoid mono-O-glycosides were recommended by characteristic fragmentation patterns. The developed method was validated using glycosides and nonglycosides from the MoNA library. The annotation accuracy rates were 96.8, 94.9, and 98.0% in negative ion mode (ESI-), positive ion mode (ESI+), and the combined ESI- & ESI+, respectively. The annotation specificity was 99.6% (ESI-), 99.6% (ESI+), and 99.2% (ESI- & ESI+). A total of 274 glycosides (including 34 acyl-glycosides) were tentatively annotated in maize by the developed method. The method enables effective and reliable annotation for plant glycosides.

L1-Norm Quantile Regression Screening Rule via the Dual Circumscribed Sphere.

l1-norm quantile regression is a common choice if there exists outlier or heavy-tailed error in high-dimensional data sets. However, it is computationally expensive to solve this problem when the feature size of data is ultra high. As far as we know, existing screening rules can not speed up the computation of the l1-norm quantile regression, which dues to the non-differentiability of the quantile function/pinball loss. In this paper, we introduce the dual circumscribed sphere technique and propose a novel l1-norm quantile regression screening rule. Our rule is expressed as the closed-form function of given data and eliminates inactive features with a low computational cost. Numerical experiments on some simulation and real data sets show that this screening rule can be used to eliminate almost all inactive features. Moreover, this rule can help to reduce up to 23 times of computational time, compared with the computation without our screening rule.

Instance elimination strategy for non-convex multiple-instance support vector machine

Multiple-instance support vector machine (SVM) is a classic and effective algorithm to address the classification of bags. However, training multiple-instance SVM is expensive. To deal with this issue, the safe screening rule is introduced to solve optimization problem, which can significantly mitigate the storage burden by identifying inactive instances. Specifically, an instance elimination strategy is designed for the inner solver of a signal concave–convex procedure (CCCP). Then, between the iterations of CCCP, due to the label of instance may change, to cope with this difficulty, a dual screening method with variational inequalities (DVI) is used to identify the part of inactive instances. To further speed up the solution efficiency, a smart dual coordinate descent method (SDCDM) is introduced for the inner solver, which skips over many updates that result in no change to the current iteration. Experiments on 35 benchmark datasets significantly exhibit the outstanding performance of the proposed strategy.

Outperformance of the pharmaceutical sector during the COVID-19 pandemic: Global time-varying screening rule development

This study demonstrates the major role played by the healthcare and pharmaceutical industries during the COVID-19 pandemic. For this purpose, it provides evidence of a better risk–return relationship in these sectors through a multivariate study of monthly frequency. A global and dynamic ratio is developed to summarize different investor profiles according to their attitude toward risk and to consider the dynamic and changing nature of the economy and financial markets. This global ratio synthesizes the information from different orders of Kappa indices and Farinelli–Tibiletti ratios into a single measure. Additionally, we verify that Principal Component Analysis allows summarizing the information contained in the initial variables from the first new component. Our findings confirm that a filtering asset screening rule strategy is relevant and necessary. In this respect, passive management of midterm equal-weighted pharmaceutical portfolios outperforms both the pure financial and the healthcare investment strategies used during the pandemic.

Target Object Detection from Unmanned Aerial Vehicle (UAV) Images Based on Improved YOLO Algorithm

Aerial image-based target object detection has several glitches such as low accuracy in multi-scale target detection locations, slow detection, missed targets, and misprediction of targets. To solve this problem, this paper proposes an improved You Only Look Once (YOLO) algorithm from the viewpoint of model efficiency using target box dimension clustering, classification of the pre-trained network, multi-scale detection training, and changing the screening rules of the candidate box. This modified approach has the potential to be better adapted to the positioning task. The aerial image of the unmanned aerial vehicle (UAV) can be positioned to the target area in real-time, and the projection relation can convert the latitude and longitude of the UAV. The results proved to be more effective; notably, the average accuracy of the detection network in the aerial image of the target area detection tasks increased to 79.5%. The aerial images containing the target area are considered to experiment with the flight simulation to verify its network positioning accuracy rate and were found to be greater than 84%. This proposed model can be effectively used for real-time target detection for multi-scale targets with reduced misprediction rate due to its superior accuracy.

Presentation and diagnosis of childhood-onset combined pituitary hormone deficiency: A single center experience from over 30 years.

SummaryBackgroundChildhood-onset combined pituitary hormone deficiency (CPHD) has a wide spectrum of etiologies and genetic causes for congenital disease. We aimed to describe the clinical spectrum and genetic etiologies of CPHD in a single tertiary center and estimate the population-level incidence of congenital CPHD.MethodsThe retrospective clinical cohort comprised 124 CPHD patients (48 with congenital CPHD) treated at the Helsinki University Hospital (HUH) Children's Hospital between 1985 and 2018. Clinical data were collected from the patient charts. Whole exome sequencing was performed in 21 patients with congenital CPHD of unknown etiology.FindingsThe majority (61%;76/124) of the patients had acquired CPHD, most frequently due to craniopharyngiomas and gliomas. The estimated incidence of congenital CPHD was 1/16 000 (95%CI, 1/11 000-1/24 000). The clinical presentation of congenital CPHD in infancy included prolonged/severe neonatal hypoglycaemia, prolonged jaundice, and/or micropenis/bilateral cryptorchidism in 23 (66%) patients; despite these clinical cues, only 76% of them were referred to endocrine investigations during the first year of life. The median delay between the first violation of the growth screening rules and the initiation of GH Rx treatment among all congenital CPHD patients was 2·2 years, interquartile range 1·2–3·7 years. Seven patients harbored pathogenic variants in PROP1, SOX3, TBC1D32, OTX2, and SOX2, and one patient carried a likely pathogenic variant in SHH (c.676G>A, p.(Ala226Thr)).InterpretationOur study suggests that congenital CPHD can occur in 1/16 000 children, and that patients frequently exhibit neonatal cues of hypopituitarism and early height growth deflection. These results need to be corroborated in future studies and might inform clinical practice.FundingPäivikki and Sakari Sohlberg Foundation, Biomedicum Helsinki Foundation, and Emil Aaltonen Foundation research grants.