Search

Abstract Multi-attribute group decision-making is an important research field in decision science, and its theories and methods have been widely applied to engineering, economics and management. However, as the information embedded volume and complexity of decision-making expand, the diversity and heterogeneity of decision-making groups present significant challenges to the decision-making process. In order to effectively address these challenges, this paper defines the concept of spherical Z-number, which is a fuzzy number that takes into account a wide range of evaluation and its reliability. Additionally, a group decision-making model in a spherical Z-number environment is proposed. First, an objective phased tracking method is used to determine expert weights, maintain the consistency in decision-making group evaluations. The gained and lost dominance score method is combined with prospect theory to integrate expert psychological behavior when facing risks. The proposed method considers both group utility and individual regret, and balances the gains and losses of various options in the decision-making process. Finally, in response to the 3R principle, the model is employed to address the shared e-bike recycling supplier selection problem and to assess the viability of the decision-making outcomes. The results demonstrate that the model is robust in the context of varying parameter configurations. Moreover, the correlation coefficients between its ranking outcomes and those of alternative methodologies are all above 0.77, and its average superiority degree is 1.121, which is considerably higher than that of other methods. Consequently, the model's effectiveness and superiority are substantiated.

Abstract Exploiting the utility of near-term quantum devices is a long-standing challenge whereas hybrid quantum machine learning emerges as a promising candidate. Here we propose a quantum-enhanced deep generative algorithm based on programmable quantum circuit-induced quantum latent codes. To validate the effectiveness of the algorithm, we conduct optical ghost imaging experiments, collecting dataset under varying physical sampling rates. Leveraging a physically enhanced loss function and a pretrained neural network, the quantum algorithm exhibits superior reconstruction performance compared to conventional algorithms. We observe the learnable quantum latent space enhances the out-of-distribution generalization capability of the pretrained model. In the context of computer vision problems, numerical results indicate that quantum latent space can increase the generation diversity of the pretrained model. Furthermore, the algorithm outperforms classical counterparts, particularly in image inpainting and colorization, by a significant margin. Our study demonstrates the utility of hybrid quantum-classical algorithms in enhancing generalization capability, highlighting the potential of near-term quantum devices in large-scale generative artificial intelligence.

BackgroundGastric cancer (GC) is one of the leading causes of cancer-related death in China, and with the extensive development of conversion therapy, the treatment of advanced unresectable gastric cancer (AUGC) patients has ushered in a new dawn. This study aimed to analyze the efficacy of conversion therapy in AUGC patients and explored the relevant factors affecting the efficacy.MethodWe collected information from GC patients who received conversion therapy from this center and designed a retrospective study.ResultsWe collected relevant clinical data from 160 patients with AUGC. A total of 120 patients who underwent routine R0 resection were identified as conversion cases. A total of 25 patients (15.6%) achieved pCR, 92 patients (57.5%) achieved objective response rate (ORR), 140 patients (87.5%) achieved disease control rate (DCR), and 20 cases (12.5%) observed tumor progression. There were 86 patients who achieved pathological downgrading, with a total downgrading rate of 53.8%. Among the 160 patients, 37 patients (23.1%) had postoperative complications of varying degrees. A total of 72 patients (45.0%) had tumor recurrence/progression at the end of follow-up. The last chemotherapy and surgery (CST) (OR = 1.046, 95% CI 1.013–1.081, p = 0.006), tumor invasion (OR = 32.096, 95% CI 5.091–202.349, p < 0.001), and distant metastasis (OR = 7.050, 95% CI 1.888–26.323, p = 0.004) were independent factors influencing the efficacy of conversion therapy.ConclusionConversion therapy may have a good therapeutic efficacy for AUGC, and some clinical factors affect the efficacy response.

Abstract Image restoration is a prominent field of research in computer vision. Restoring broken paintings, especially ancient Chinese artworks, is a significant challenge for current restoration models. The difficulty lies in realistically reinstating the intricate and delicate textures inherent in the original pieces. This process requires preserving the unique style and artistic characteristics of the ancient Chinese paintings. To enhance the effectiveness of restoring and preserving traditional Chinese paintings, this paper presents a framework called Sketch-Guided Restoration Generative Adversarial Network, termd SGRGAN. The framework employs sketch images as structural priors, providing essential information for the restoration process. Additionally, a novel Focal block is proposed to enhance the fusion and interaction of textural and structural elements. It is noteworthy that a BiSCCFormer block, incorporating a Bi-level routing attention mechanism, is devised to comprehensively grasp the structural and semantic details of the image, including its contours and layout. Extensive experiments and ablation studies on MaskCLP and Mural datasets demonstrate the superiority of the proposed method over previous state-of-the-art methods. Specifically, the model demonstrates outstanding visual fidelity, particularly in the restoration of landscape paintings. This further underscores its efficacy and universality in the realm of cultural heritage preservation and restoration.

Abstract The large-scale multi-attribute group decision-making (LSMAGDM) problem has become a hot research topic in the field of decision science. An R-numbers large-scale multi-attribute group decision-making (R-LSMAGDM) model is proposed to be constructed in this paper based on the advantages of R-numbers in capturing risks. First, the most commonly used clustering method, k-means, is introduced to determine the sub-groups. Then, a new sub-group weighting determination model is constructed by considering sub-group size and sub-group entropy. Next, we also build an optimized consensus-reaching model by improving the calculation method of the mean value. Then, the R-numbers weighted Hamy mean (RNWHM) operator is proposed to aggregate the sub-group information. In addition, the logarithmic percentage change-driven objective weighting (LOPCOW) method and the compromise ranking of alternatives from distance to ideal solution (CRADIS) method are used for attribute weighting calculation and alternative ranking, respectively. Finally, the effectiveness of the model is verified by an application example of hydrogen fuel cell logistics path selection.

Nonnegative Matrix Factorization (NMF) is one of the most popular feature learning technologies in the field of machine learning and pattern recognition. It has been widely used and studied in the multi-view clustering tasks because of its effectiveness. This study proposes a general semi-supervised multi-view nonnegative matrix factorization algorithm. This algorithm incorporates discriminative and geometric information on data to learn a better-fused representation, and adopts a feature normalizing strategy to align the different views. Two specific implementations of this algorithm are developed to validate the effectiveness of the proposed framework: Graph regularization based Discriminatively Constrained Multi-View Nonnegative Matrix Factorization (GDCMVNMF) and Extended Multi-View Constrained Nonnegative Matrix Factorization (ExMVCNMF). The intrinsic connection between these two specific implementations is discussed, and the optimization based on multiply update rules is presented. Experiments on six datasets show that the effectiveness of GDCMVNMF and ExMVCNMF outperforms several representative unsupervised and semi-supervised multi-view NMF approaches.

E-governance is capable of driving the transition of government from an administrative power allocation of public resources to a citizen-service-oriented governance model. However, during this transformation, various governmental bodies often face the challenge of "data silos" caused by factors such as cross-regional, cross-business, and cross-departmental operations. Without altering the existing information platforms, transparent access technology serves as a key solution for data access within egovernance systems. It enables convenient access to information resources stored in different mediums and formats, facilitating the sharing and consolidation of information and data within and between governmental departments, thereby addressing the issue of "data silos" and enhancing the comprehensive service capabilities of e-governance. This paper firstly provides an overview of the concept, levels, characteristics, and application scenarios of transparent access to government big data. Secondly, it conducts a comprehensive comparative analysis of transparent access technologies in the context of cloud computing and big data. Lastly, based on the requirements of various transparent access technologies and the application of transparent access to government big data, this paper proposes a visionary framework for transparent access to government big data based on cross-domain semantics and channel coupling. This framework includes modules for cross-domain semantic interoperability, coupling of heterogeneous information channels based on knowledge graph, and tracing of multi-source heterogeneous data, aiming to provide innovative solutions for achieving transparent access to government big data.

Abstract Petroleum is the main power of automobiles, yet the environmental pollution it brings has caused globally. The promotion of new energy vehicles (NEVs) is effective. Even though the governments have introduced various supports for the popularization of NEVs, many enterprises and individuals continue to drive gas vehicles as they are poorly informed about them and cannot choose the right products. This paper proposes a categorization and ranking model of NEVs over three-way decisions (TWD) to tackle the problem. Picture fuzzy sets serve as a valid tool to describe uncertainty and fuzziness information. At first, the attribute weights are ascertained with the indifference threshold-based attribute ratio analysis (ITARA) method improved by the picture fuzzy distance. Then, the public weight picture fuzzy data envelopment analysis (DEA) which is enhanced by the picture fuzzy aggregation operator is employed for the conditional probability of decision units. Third, updated picture fuzzy TWD is utilized for the categorization and ranking. Fourth, the enterprise NEVs selection case is tested for the model. Later, comparative analysis with the classical methods illustrates that the proposed method yields accurate and reliable decisions. Finally, the robustness of the model is demonstrated by sensitivity analysis.

From the Late Carboniferous to the end of Middle Permian, the most intense glacial event in history of Phanerozoic occurred and it is also the last transition period from icehouse stage to greenhouse stage in geological history. In this paper, the paleotemperature and paleoclimate in the first member of the Middle Permian Maokou Formation (Mao-1) were restored by means of thin section identification, scanning electron microscope, major elements, trace elements, carbon isotopes, oxygen isotopes and strontium isotopes, and the major climate evolution process and its driving mechanism in the Middle Permian were discussed. The results show that the biological assemblage type (mainly composed of foraminifera, brachiopods and molluscs, without reef-building organisms or calcareous green algae) and rock structure characteristics (supported by bioclastic and plaster, without oolitic or other non-skeletal particles) of the limestone-marl alternations in the Mao-1 member are similar to those of the international typical cool-water carbonate. During the limestone sedimentary period, the paleotemperature of seawater was 3.72 to 12.38 ℃ (8.15 ℃ in average, δ18O standard) or 13.79 to 14.28 ℃ (13.90 ℃ in average, ω(Mg)/ ω(Ca) standard), while during the marl sedimentary period, the paleotemperature of seawater was 7.00 to 14.24 ℃ (10.97 ℃ in average, δ18O standard) and 13.82 to 15.41 ℃ (14.27 ℃ in average, ω(Mg)/ω(Ca) standard). The paleotemperature changes in the sedimentary period of Mao-1 Member was mainly driven by the short eccentricity cycle of Milankovitch. The cyclical change of the short eccentricity was the driving mechanism of paleotemperature and paleoclimate cycle changes.

Abstract Demonstrating the practical advantage of quantum computation remains a long-standing challenge whereas quantum machine learning becomes a promising application that can be resorted to. In this work, we investigate the practical advantage of quantum machine learning in ghost imaging by overcoming the limitations of classical methods in blind object identification and imaging. We propose two hybrid quantum-classical machine learning algorithms and a physical-inspired patch strategy to allow distributed quantum learning with parallel variational circuits. In light of the algorithm, we conduct experiments for imaging-free object identification and blind ghost imaging under different physical sampling rates. We further quantitatively analyze the advantage through the lens of information geometry and generalization capability. The numerical results showcase that quantum machine learning can restore high-quality images but classical machine learning fails. The advantage of identification rate are up to 10% via fair comparison with the classical machine learning methods. Our work explores a physics-related application capable of practical quantum advantage, which highlights the prospect of quantum computation in the machine learning field.

Objective It is clinically important for pedicle screws to be placed quickly and accurately. Misplacement of pedicle screws results in various complications. However, the incidence of complications varies greatly due to the different professional titles of physicians and surgical experience. Therefore, physicians must minimize pedicle screw dislocation. This study aims to compare the three nail placement methods in this study, and explore which method is the best for determining the anatomical landmarks and vertical trajectories. Methods This study involved 70 patients with moderate idiopathic scoliosis who had undergone deformity correction surgery between 2018 and 2021. Two spine surgeons used three techniques (preoperative computed tomography scan [CTS], visual inspection‐X‐freehand [XFH], and intraoperative detection [ID] of anatomical landmarks) to locate pedicle screws. The techniques used include visual inspection for 287 screws in 21 patients, preoperative planning for 346 screws in 26 patients, and intraoperative probing for 309 screws in 23 patients. Observers assessed screw conditions based on intraoperative CT scans (Grade A, B, C, D). Results There were no significant differences between the three groups in terms of age, sex, and degree of deformity. We found that 68.64% of screws in the XFH group, 67.63% in the CTS group, and 77.99% in the ID group were placed within the pedicle margins (grade A). On the other hand, 6.27% of screws in the XFH group, 4.33% in the CTS group, and 6.15% in the ID group were considered misplaced (grades C and D). The results show that the total amount of upper thoracic pedicle screws was fewer, meanwhile their placement accuracy was lower. The three methods used in this study had similar accuracy in intermediate physicians (P > 0.05). Compared with intermediate physicians, the placement accuracy of three techniques in senior physicians was higher. The intraoperative detection group was better than the other two groups in the good rate and accuracy of nail placement (P

Abstract Dynamic multi-attribute group decision-making (DMAGDM) is a widespread practice in which evaluations are provided by multiple decision-makers at various times and early evaluations impact later evaluations. Additionally, attributes and alternatives can be added or removed over time. An R-numbers DMAGDM method is developed based on the advantages of R-numbers in capturing risks. This paper introduces the R-numbers Einstein weighted averaging (RNEWA) operator and R-numbers weighted Einstein geometric (RNEWG) operator, which are distinct from conventional algebraic operations, and examines their properties. Moreover, an expert weight determination model is constructed using the similarity measure of R-numbers. The attribute weight determination model in the R-numbers environment is also proposed with the method based on the criteria removal effects method (MEREC). A static rating calculation model, which utilizes the combination compromise solution (CoCoSo) method in the R-numbers environment, is built using the RNEWA operator and RNEWG operator. Furthermore, a new dynamic rating calculation model is proposed which does not require storage of all decision information over time. Finally, the applicability and effectiveness of the R-numbers DMAGDM method is demonstrated through a case study on supply chain risk assessment of manufacturing enterprises.

Abstract The purpose of this paper is to study the multi-attribute decision-making problem under the fuzzy picture environment. First, a method to compare the pros and cons of picture fuzzy numbers (PFNs) is introduced in this paper. Second, the correlation coefficient and standard deviation (CCSD) method is used to determine the attribute weight information under the picture fuzzy environment regardless of whether the attribute weight information is partially unknown or completely unknown. Third, the ARAS and VIKOR methods are extended to the picture fuzzy environment, and the proposed PFNs comparison rules are also applied in the PFS-ARAS and PFS-VIKOR methods. Fourth, the problem of green supplier selection in a picture-ambiguous environment is solved by the method proposed in this paper. Finally, the method proposed in this paper is compared with some methods and the results are analyzed.

Abstract Quantum simulation is a technology of using controllable quantum systems to study new quantum phases of matter. Certification for quantum simulators is a challenging problem whereas identification and properties estimation are two crucial approaches that can be resorted to. In this work, we propose Ab initio end-to-end machine learning certification protocol briefly named MLCP. The learning protocol is trained with a million-level size of randomized measurement samples without relying on the assistance of quantum tomography. In the light of MLCP, we can identify different types of quantum simulators to observe their distinguishability hardness. We also predict the physical properties of quantum states evolved in quantum simulators such as entanglement entropy and maximum fidelity. The impact of randomized measurement samples on the identification accuracy is analyzed to showcase the potential capability of classical machine learning on quantum simulation results. The entanglement entropy and maximum fidelity with varied subsystem partitions are also estimated with satisfactory precision. This work paves the way for large-scale intelligent certification of quantum simulators and can be extended onto an artificial intelligence center to offer easily accessible services for local quantum simulators in the noisy intermediate-size quantum (NISQ) era.

Abstract Direction‐of‐arrival (DOA) estimation is a fundamental functionality of sensor array systems. Previous methods only consider element failure or array imperfections. In fact, the co‐existence of element failure and array imperfections is more in line with the realistic operation conditions of an array system. However, the performance of previous methods degrades significantly in the presence of both array imperfections and element failure. To deal with this problem, a robust DOA estimation method is proposed based on a deep neural network (DNN). The proposed DNN consists of a denoising autoencoder (DAE) and a parallel network. The DAE can restore damaged array signals to “non‐corrupted” signals, and the parallel network is able to process signals with different levels of loss to improve DOA estimation accuracy. At the training stage, array imperfections are modelled as a spherical distribution, and the training samples are extracted under this distribution to improve the generalisation capability of the proposed network to various array imperfections. In addition, the authors propose to estimate the spatial spectrum in a virtual array beamspace, which can reduce the computational complexity as well as signal‐to‐noise ratio resolution threshold, and further enhance the robustness to array imperfections. Numerical results show that the proposed DOA estimation approach works well in the presence of both array imperfections and element failure.

The R-set is modeled by considering the risks associated with sources of information and future events. In the current work, there is no research on the aggregation operator of the R-set, and no in-depth research on its multi-attribute decision-making (MADM) methods. In order to extend the application of R-set more widely, this paper aims to propose a new risk-based fuzzy complex proportional assessment (R-COPRAS) method. First, the R-set Bonferroni mean (RBM) operator and the R-set geometric Bonferroni mean (RGBM) operator are investigated. Then, the correlation properties of these operators are studied. Based on these two operators, COPRAS methods in R-set environment are proposed, which consider the decision risk and can provide stable decision results. Ultimately, a smart mine safety assessment project is proposed to test and check the accuracy and effectiveness of the proposed R-COPRAS methods.

Abstract Parameter estimation is a pivotal task, where quantum technologies can enhance precision greatly. We investigate the time-dependent parameter estimation based on deep reinforcement learning, where the noise-free and noisy bounds of parameter estimation are derived from a geometrical perspective. We propose a physical-inspired linear time-correlated control ansatz and a general well-defined reward function integrated with the derived bounds to accelerate the network training for fast generating quantum control signals. In the light of the proposed scheme, we validate the performance of time-dependent and time-independent parameter estimation under noise-free and noisy dynamics. In particular, we evaluate the transferability of the scheme when the parameter has a shift from the true parameter. The simulation showcases the robustness and sample efficiency of the scheme and achieves the state-of-the-art performance. Our work highlights the universality and global optimality of deep reinforcement learning over conventional methods in practical parameter estimation of quantum sensing.

Abstract Background Fibrosis is an important factor and process of ligamentum flavum hypertrophy. The expression of phosphodiesterase family (PDE) is related to inflammation and fibrosis. This article studied the expression of PDE in hypertrophic ligamentum flavum fibroblasts and investigated whether inhibition of PDE4 activity can play an anti-fibrotic effect. Methods Samples of clinical hypertrophic ligamentum flavum were collected and patients with lumbar disc herniations as a control group. The collagenase digestion method is used to separate fibroblasts. qPCR is used to detect the expression of PDE subtypes, type I collagen (Col I), type III collagen (Col III), fibronectin (FN1) and transforming growth factor β1 (TGF-β1). Recombinant TGF-β1 was used to stimulate fibroblasts to make a fibrotic cell model and treated with Rolipram. The morphology of the cells treated with drugs was observed by Sirius Red staining. Scratch the cells to observe their migration and proliferation. WB detects the expression of the above-mentioned multiple fibrotic proteins after drug treatment. Finally, combined with a variety of signaling pathway drugs, the signaling mechanism was studied. Results Multiple PDE subtypes were expressed in ligamentum flavum fibroblasts. The expression of PDE4A and 4B was significantly up-regulated in the hypertrophic group. Using Rolipram to inhibit PDE4 activity, the expression of Col I and TGF-β1 in the hypertrophic group was inhibited. Col I recovered to the level of the control group. TGF-β1 was significantly inhibited, which was lower than the control group. Recombinant TGF-β1 stimulated fibroblasts to increase the expression of Col I/III, FN1 and TGF-β1, which was blocked by Rolipram. Rolipram restored the increased expression of p-ERK1/2 stimulated by TGF-β1. Conclusion The expressions of PDE4A and 4B in the hypertrophic ligamentum flavum are increased, suggesting that it is related to the hypertrophy of the ligamentum flavum. Rolipram has a good anti-fibrosis effect after inhibiting the activity of PDE4. This is related to blocking the function of TGF-β1, specifically by restoring normal ERK1/2 signal.

Abstract Image classification is one of the mainstream tasks of computer vision. However, the most existing methods use labels of the same granularity level for training. This leads to ignoring the hierarchy that may help to differentiate different visual objects better. Embedding hierarchical information into the convolutional neural networks (CNNs) can effectively regulate the semantic space and thus reduce the ambiguity of prediction. To this end, a multi‐task learning framework, named as Hierarchical Bilinear Convolutional Neural Network (HB‐CNN), is developed by seamlessly integrating CNNs with multi‐task learning over the hierarchical visual concept structures. Specifically, the labels with a tree structure are used as the supervision to hierarchically train multiple branch networks. In this way, the model can not only learn additional information (e.g. context information) as the coarse‐level category features, but also focus the learned fine‐level category features on the object properties. To smoothly pass hierarchical conceptual information and encourage feature reuse, a connectivity pattern is proposed to connect features at different levels. Furthermore, a bilinear module is embedded to generalise various orderless texture feature descriptors so that our model can capture more discriminative features. The proposed method is extensively evaluated on the CIFAR‐10, CIFAR‐100, and ‘Orchid’ Plant image sets. The experimental results show the effectiveness and superiority of our method.

Purpose – This paper aims to systematically expound the theory and development background of supply chain finance and blockchain, design a railway freight supply chain financial platform based on blockchain, determine the risk management system and business support system of supply chain finance business and analyze the value generated by the combination of supply chain finance business and blockchain. Design/methodology/approach – Investigation and research method; Prototype method; Model method; Value analysis. Findings – The business model integrating supply chain finance and blockchain technology will bring great changes to freight industry. The development of supply chain finance is beneficial to the healthy development of the core participants of railway freight transport business and its upstream and downstream ecosystems. It links commerce, logistics, warehousing and financial services together and builds an industry-integrated ecological service platform through information technology platform and supporting system, taking data as the basis and combining information technology such as blockchain as innovative means. Originality/value – This paper will provide important reference value for related research. This paper innovatively designs the supply chain financial platform of freight transportation industry-integrating blockchain technology and analyzes its business model, technical system, risk management and control system and value system in detail, which will provide technical support for the innovative reform of freight information technology and realize the stable and high-speed development of freight logistics informationization.

Detecting communities in complex networks has been one of the most popular research areas in recent years. There have been many community detection algorithms proposed to date. However, the local information (cliques) of communities and the search efficiency of algorithm have not been considered both in previous studies. In this paper, we propose a novel local expansion algorithm for detecting overlapping communities based on cliques. The algorithm draws on the assumption that cliques are the core of communities, as the clique takes into account the local characteristics of the community. The proposed algorithm adopts a single node with the maximum density as an initial community to prevent the formation of a large number of near-duplicate community structures, which improves the search efficiency of the algorithm. In many experiments using computer-generated and real-world networks, the proposed algorithm based on this idea verifies that the algorithm is able to detect overlapping communities effectively. The experiment yields better community uncover results, and the time efficiency and the complexity of algorithm are also satisfactory.

In this paper, a novel loss function is proposed to measure the correlation among different learning tasks and select useful feature components for each classification task. Firstly, the knowledge map we proposed is used for organizing the affiliation relationship between objects in natural world. Secondly, a novel loss function-orthogonality loss is proposed to make the deep features more discriminative by removing useless feature components. Furthermore, in order to prevent the extracted feature maps from being too divergent and causing over-fitting which will reduce network performance, this paper also added the orthogonal distribution regularization term to constrain the distribution of network parameters. Finally, the proposed orthogonality loss is applied in a multi-task network structure to learn more discriminative deep feature, and also to evaluate the validity of the proposed loss function.The results show that compared with the traditional deep convolutional neural network and a multi-task network without orthogonality loss, the multi -task based orthogonality loss is significantly better than the other two types of networks on image classification.

Neurodegenerative diseases are common progressive nervous system disorders that show intricate clinical patterns. The gait fluctuations reflect the physiology and pathologic alterations in the locomotor control system. Using gait fluctuations for disease state evaluation is an essential way for clinical trials and healthcare monitoring. The classification of gait fluctuations helps improve the life quality and enhance clinical diagnosis ability in neuro-degenerative patients. In this work, we firstly embed the time series of multiple gait fluctuations into the phase space. Then we use persistent homology to extract the topological signatures of barcodes. Together with a random forest classifier, we proposed a topological motion analysis (TMA) framework to analyze the gait fluctuations. Further, we proposed a comprehensive comparison study using the TMA framework in the neuro-degenerative classification tasks for stance-, stride-, and swing-based gait fluctuations. In the tasks of comparing amyotrophic lateral sclerosis (ALS), Huntington's disease (HD), and Parkinson's disease (PD) to the healthy control (HC) group, the best-achieved AUC scores were 0.9135, 0.9906 and 0.9667 respectively, which show the effectiveness of TMA framework. In summary, our study proposed a TMA framework towards gait fluctuations classification in the neuro-degenerative analysis tasks. The proposed method shows promising clinical application value in earlier interventions and state monitoring for neurodegenerative patients.

In this paper, a novel hierarchical deep network is proposed to combine the deep convolutional neural network and the feature selection-based tree classifier efficiently for image classification. First, the concept ontology is built for organizing large-scale image classes hierarchically in a coarse-to-fine fashion. Second, a novel selective orthogonal algorithm is proposed to make sure deep features extracted for each level classifiers more in line with the requirements of different classification tasks. Also, the role of useful feature components in multi-level deep features are improved. The experimental results on three datasets show that adding a feature selection module in a hierarchical deep network can perform better performance in large-scale image classification.

As third-party logistics providers occupy an increasingly important position in the operation strategies of various enterprises, choosing a partner that suits them and can bring benefits to them has become an important decision-making issue for each enterprise. A lot of studies have been done with classical fuzzy methods, but there are relatively few studies that consider information reliability. The emergence of Z-numbers makes up for this deficiency, including the restriction on the evaluation object and the corresponding degree of confidence. In this article, making decisions based on Z-numbers can better transform language values into fuzzy numbers. By integrating the existing Multi-Attributive Border Approximation area Comparison (MABAC) method and Z-numbers, a new method for the selection of third-party logistics providers is provided. Finally, the feasibility and effectiveness of the approach are verified by comparing with the classical Multi Criteria Decision Making methods. The Z-number is a relatively new concept that can flexibly represent the confidence of information, which will be a relatively important research direction in the future.

This paper studies the similarities and the differences between natural aggregate concrete (NAC) beams and recycled aggregate concrete (RAC) beams in terms of concrete material properties, bearing capacity, and crack resistance and further proposes correction formulas for cracking moment calculation of RAC beams that consider different recycled aggregate substitution ratios. First, the basic mechanical properties (e.g., cube compressive strength, axial compressive strength, and elastic modulus) of RAC blocks were tested; second, comparative bending tests of RAC beams and natural aggregate concrete beams were conducted, in order to obtain the cracking moments, yield bending moments, ultimate bending moments, mid-span deflections and crack development forms; finally, numerical simulations were performed to investigate the cracking performances of RAC beams. Through analyzing the experimental results, it is found that the crack development pattern of RAC beams resembles that of natural aggregate concrete beams, while the cracking performance of RAC beams significantly deviates from that of NAC beams. Finally, the correction formulas for the cracking moment calculation of RAC beams are proposed based on the numerical simulation results and verified by data from other researchers, which can be regarded as a correction for the cracking moment calculation formulas in the current code for RAC beams with respect to varied recycled concrete aggregate substitution ratios.

Investigating quantum advantage in the NISQ era is a challenging problem whereas quantum machine learning becomes the most promising application that can be resorted to. However, no proposal has been investigated for arguably challenging inverse reinforcement learning to demonstrate the potential advantage. In this work, we propose a hybrid quantum–classical inverse reinforcement learning algorithm based on the variational quantum circuit with the generative adversarial framework. We find an important connection between the quantum gradient anomaly and the performance degradation, which suggest a gradient clipping strategy to stabilize the training process. In light of the algorithm, we study three classic control problems and the Hamiltonian parameter estimation in quantum sensing with shallow quantum circuits. The numerical results showcase that the control-enhanced quantum sensor can saturate quantum Cramér-Rao bound only with a single variational layer, empirically demonstrating a parameter complexity advantage over the classical learning control. The proposed generative adversarial reinforcement learning algorithm achieves state-of-the-art performance in classical and quantum sensor control in terms of required number of parameters.

The image fusion method based on sparse representation in the single‐scale image domain has produced better fusion results than the classic methods based on multi‐scale analysis nowadays. However, due to the limited number of dictionary atoms, it is difficult to provide an accurate description for image details in the sparse‐representation‐based image fusion methods, and it requires a lot of time. A novel dictionary is constructed with non‐subsampled contourlet transform and sparse representation by using the proposed simultaneous strategy. Then the novel dictionary could combine the sparsity attribute of the learning dictionary with a multi‐scale feature of non‐subsampled contourlet transform. Moreover, the simultaneous strategy is combined with this novel dictionary so that sparse coefficients can be represented with the same dictionary atoms and thus they can be compared in a reasonable and accurate way. Finally, the image fusion method along with this novel dictionary is proposed and named non‐subsampled contourlet transform (NSCT)–simultaneous sparse representation (SSR). Experimental results show that the proposed fusion method NSCT–SSR, with its more excellent fusion effect and better anti‐noise capability, outperforms the existing fusion methods, which are based on both multi‐scale domain and sparse representation in the single‐scale image domain.

Safety accidents caused by Lithium-ion (Li-ion) batteries are numerous in recent years. Therefore, more and more attention has been drawn to the Remaining Useful Life (RUL) prediction and health status monitoring for Li-ion batteries. This paper proposes a deep learning method that combines the Forgetting Online Sequential Extreme Learning Machine (FOS-ELM) with the Hybrid Grey Wolf Optimizer (HGWO) algorithm and attention mechanism for the Prognostic and Health Management (PHM) of Li-ion battery. First, we use the Variational Mode Decomposition (VMD) to denoise the raw data before the training. Then the key parameters optimization of the FOS-ELM model based on the HGWO algorithm is introduced. Finally, we apply the attention mechanism to further improve the accuracy of the algorithm. Compared with traditional neural network methods, the method proposed in this paper has higher efficiency and accuracy.

Failure mode and effect analysis (FMEA) is a powerful risk assessment tool to eliminate the risk and improve the reliability. In this article, a novel risk priorization model based on extended TODIM (an acronym in Portuguese of interactive and multiple attribute decision-making) method under interval-valued neutrosophic environment is proposed. Firstly, the interval-valued neutrosophic sets (IVNSs) are adopted to deal with uncertainty and indeterminate information. Secondly, in order to obtain objective weights of risk factors, integration of the new similarity degree and entropy measures are applied for risk factor weighting. Moreover, the extended TODIM method is present to reduce fuzziness in process of decision-making by using an improved score function, and it also attach importance to the psychological behavior of team members, which can get a more reasonable ranking. Finally, a numerical case of steel company is provided to illustrate the feasibility of the FMEA model, and a comparison analysis with other conventional methods are further performed to indicate its effectiveness.

Exploring the spatiotemporal characteristics of park visitors and the “push and pull” factors that shape this mobility is critical to designing and managing urban parks to meet the demands of rapid urbanization. In this paper, 56 parks in Shenzhen were studied in 2019. First, cell phone signaling data were used to extract information on visitors’ departure locations and destination parks. Second, the bivariate Moran’s I and bivariate local Moran’s I (BiLISA) methods were used to identify the statistical correlation between the factors of the built environment and the park recreation trips. Finally, linear regression models were constructed to quantify the factors influencing the attractiveness of the park. Our study showed the following: (1) Recreation visitors at large parks varied significantly among population subgroups. Compared with younger adults, teenagers and older adults traveled lower distances and made fewer trips, and in particular, older adults of different genders differed significantly in park participation. (2) Recreational trips in large parks were related to the functional layout of the built environment around their residence. In areas with rich urban functions (e.g., southern Shenzhen), trips to large parks for leisure are more aggregated. (3) The findings reinforce the evidence that remote sensing data for urban vegetation can be an effective factor in characterizing park attractiveness, but the explanatory power of different vegetation data varies widely. Our study integrated the complementary human activity and remote sensing data to provide a more comprehensive understanding of urban park use and preferences. This will be important for future park planning.

There exist a great amount of uncertain factors in actual engineering. In order to involve these uncertain factors in analytical model, they have been expressed as the convex variables. In addition, the convex model was further classified into the hyper-ellipsoidal model and the interval model. After pointing out the intrinsic difference between these two kinds of models, the principle for applying which one of the models within the analysis has been indicated according to the available testing points. After standardizing the convex variables, the difference and relation between these two models for the optimization and solution process have been presented. With the analysis mentality available from the hyper-ellipsoidal model, the basic method about the robust optimization for the interval model was emphasized. After classification of the interval variables within the optimization process, the characteristics of the robust optimization were highlighted with different constraint conditions. Using the target-performance-based analytical scheme, the algorithm, the solution step and the convergence criteria for the robust optimization have been also presented with only one reliability index. Numerical examples and engineering problems were used to demonstrate the effectiveness and correctness of the proposed approach. Keywords: Robust optimization, Non-probabilistic reliability, Interval model, Hyper-ellipsoidal model, Probabilistic index

The choice of green suppliers involves a large amount of inaccurate, incomplete, and inconsistent information, and the single-valued triangular Neutrosophic number that is an extension of the single-valued Neutrosophic number can effectively handle such problems. Considering the advantages of the single-valued triangular Neutrosophic number, this paper proposes a new aggregate operator to solve the problem of multi-criteria decision making. The new aggregate operator takes into account the priority relationship and the interrelationship between the criteria. To make the new aggregate operator more flexible, this paper introduces the Dombi operations. This paper combines the Dombi operations with the prioritized average operator and the Bonferroni mean operator to propose the single-valued triangular Neutrosophic Dombi prioritized normalized Bonferroni mean (SVTNDPNBM) operator. Finally, the SVTNDPNBM operator is applied to the problem of the green supplier selection, which proves its feasibility and stability.

Background: Meconopsis horridula Hook (M. horridula) has been used as a traditional Tibetan medicine to relieve heat and pain as well as mobilize static blood, and it is recognized as a good treatment for bruises. This study is the first trial to evaluate the tumor inhibitory activity of M. horridula extract and its underlying mechanism in the hope of providing evidence to support the anticancer function of M. horridula. Methods and Results: M. horridula extract was cytotoxic to L1210 cells in a dose- and time-dependent manner. SEM (scanning electron microscope) observation revealed obvious morphological changes in L1210 cells after M. horridula treatment. Flow cytometry analysis demonstrated that the extract dose-dependently induced early apoptosis. Additional apoptosis parameters, such as alterations in nuclear morphology and DNA damage, were also observed. Furthermore, M. horridula treatment induced G2/M arrest. M. horridula treatment significantly increased reactive oxygen species (ROS) production, suggesting that ROS are a key factor in M. horridula-induced apoptosis. Volatile constituent detection found 15 abundant chemicals in M. horridula, which may contribute to its anticancer effect. Conclusion: In conclusion, M. horridula extract induced L1210 cell apoptosis and inhibited proliferation through G2/M phase arrest, and ROS were involved in the process.

Objectives: Meconopsis integrifolia (M. integrifolia) is one of the most popular members in Traditional Tibetan Medicine. This study aimed to investigate the anticancer effect of M. integrifolia and to detect the underlying mechanisms of these effects. Methods: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay and trypan blue assay were used to evaluate the cytotoxicity of M. integrifolia. Changes in cell nuclear morphology and reactive oxygen species (ROS) level were observed by fluorescent microscopy. Apoptosis ratio, DNA damage and mitochondrial membrane potential (MMP) loss were analyzed by flow cytometry. Western blotting assay was adopted to detect the proteins related to apoptosis. Immunofluorescence was used to observe the release of cytochrome C. Results: The obtained data revealed that M. integrifolia could significantly inhibit K562 cell viability, mainly by targeting apoptosis induction and cell cycle arrest in G2/M phase. Collapse in cell morphology, chromatin condensation, DNA damage and ROS accumulation were observed. Further mechanism detection revealed that mitochondrion might be a key factor in M. integrifolia-induced apoptosis. Conclusions: M. integrifolia could induce mitochondria mediated apoptosis and cell cycle arrest in G2/M phase with little damage to normal cells, suggesting that M. integrifolia might be a potential and efficient anticancer agent that deserves further investigation.

5-Fluorouracil (5-Fu) has been widely used as a first-line drug for colorectal cancer (CRC) treatment but limited by drug resistance and severe toxicity. The chemo-sensitizers that augment its efficiency and overcome its limitation are urgently needed. Gypenosides (Gyp), the main components from Gynostemma pentaphyllum (Thunb.) Makino, has shown potential anti-tumor property with little side-effect. Here, we carefully explored the chemo-sensitization of Gyp to potentiate the anti-tumor effect of 5-Fu in vitro and in vivo.3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltertrazolium bromide tetrazolium assay and colony formation test reveal that Gyp could significantly enhance the 5-Fu-caused SW-480,SW-620 and Caco2 cells viability loss. Calcusyn analysis shows that Gyp acts synergistically with 5-Fu. Annexin V-PE/7-AAD staining indicates 5-Fu + Gyp could induce SW-480 cell apoptosis. The activations of caspase 3, caspase 9 and poly (ADP-ribose) polymerase (PARP) were involved in the process. Gyp was also found to up-regulate 5-Fu-caused phospho-p53 expression and thus augment 5-Fu-induced G0/G1 phase arrest. Gyp elevated intracellular ROS level, significantly enhanced 5-Fu-triggered DNA damage response as evidenced by flow cytometry, comet assay and the expression of Ser139-Histone H2A.X. Inhibition of ROS and p53 respectively reversed the cell death induced by 5-Fu + Gyp, suggesting the key roles of ROS and p53 in the process. Moreover, 5-Fu and Gyp in combination exhibits much superior tumor volume and weight inhibition on CT-26 xenograft mouse model in comparison to 5-Fu or Gyp alone. Immunohistochemistry analysis suggests the combinations greatly suppressed tumor proliferation. Preliminary toxicological results show that 5-Fu + Gyp treatment is relatively safe.As a potential chemo-sensitizer, Gyp displays a splendid synergistic effect with 5-Fu to inhibit cancer cell proliferation and tumor growth. By using 5-Fu and Gyp in combination would be a promising therapeutic strategy for CRC treatment.

Matrix factorization based methods have widely been used in data representation. Among them, Non-negative Matrix Factorization (NMF) is a promising technique owing to its psychological and physiological interpretation of spontaneously occurring data. On one hand, although traditional Laplacian regularization can enhance the performance of NMF, it still suffers from the problem of its weak extrapolating ability. On the other hand, standard NMF disregards the discriminative information hidden in the data and cannot guarantee the sparsity of the factor matrices. In this paper, a novel algorithm called ℓ 2 , 1 norm and Hessian Regularized Non-negative Matrix Factorization with Discriminability (ℓ 2 , 1 HNMFD), is developed to overcome the aforementioned problems. In ℓ 2 , 1 HNMFD, Hessian regularization is introduced in the framework of NMF to capture the intrinsic manifold structure of the data. ℓ 2 , 1 norm constraints and approximation orthogonal constraints are added to assure the group sparsity of encoding matrix and characterize the discriminative information of the data simultaneously. To solve the objective function, an efficient optimization scheme is developed to settle it. Our experimental results on five benchmark data sets have demonstrated that ℓ 2 , 1 HNMFD can learn better data representation and provide better clustering results.