Search

A transition-metal-free strategy regarding an iodine–sodium percarbonate catalysis to achieve the ortho-aminomethylation of phenols in aqueous media has been developed. This method can effectively broaden a wide range of phenols, tolerate sensitive functional groups, and achieve the late-stage functionalization of ten functional molecules that contain phenolic structures. [ABSTRACT FROM AUTHOR]

Conventional machine learning studies generally assume close-environment scenarios where important factors of the learning process hold invariant. With the great success of machine learning, nowadays, more and more practical tasks, particularly those involving open-environment scenarios where important factors are subject to change, called open-environment machine learning in this article, are present to the community. Evidently, it is a grand challenge for machine learning turning from close environment to open environment. It becomes even more challenging since, in various big data tasks, data are usually accumulated with time, like streams , while it is hard to train the machine learning model after collecting all data as in conventional studies. This article briefly introduces some advances in this line of research, focusing on techniques concerning emerging new classes, decremental/incremental features, changing data distributions and varied learning objectives, and discusses some theoretical issues. [ABSTRACT FROM AUTHOR]

Current neural networks are mostly built on the MP model, which usually formulates the neuron as executing an activation function on the real-valued weighted aggregation of signals received from other neurons. This letter proposes the flexible transmitter (FT) model, a novel bio-plausible neuron model with flexible synaptic plasticity. The FT model employs a pair of parameters to model the neurotransmitters between neurons and puts up a neuron-exclusive variable to record the regulated neurotrophin density. Thus, the FT model can be formulated as a two-variable, two-valued function, taking the commonly used MP neuron model as its particular case. This modeling manner makes the FT model biologically more realistic and capable of handling complicated data, even spatiotemporal data. To exhibit its power and potential, we present the flexible transmitter network (FTNet), which is built on the most common fully connected feedforward architecture taking the FT model as the basic building block. FTNet allows gradient calculation and can be implemented by an improved backpropagation algorithm in the complex-valued domain. Experiments on a broad range of tasks show that FTNet has power and potential in processing spatiotemporal data. This study provides an alternative basic building block in neural networks and exhibits the feasibility of developing artificial neural networks with neuronal plasticity. [ABSTRACT FROM AUTHOR]

Background: Wilson's disease (WD) is one of the few hereditary diseases that can be successfully treated with medicines. We conduct this survey research to assess treatment persistence among patients with WD and try to identify what factors affect the treatment persistence. Methods: We employed WeChat which is the most popular social software in China to carry out this anonymous questionnaire research. The questionnaire included medication adherence scale. We also collected available medical records related to demographic and clinical characteristics. All the patients were divided into group of persistence with drug treatment (PDT) and nonpersistence with drug treatment (n‐PDT). Results: We collected 242 qualified questionnaires. Only 66.5% of patients were PDT during the mean 12.6 years of follow‐up. In PDT group, better outcomes were observed: improvement (78.3%) and no change (16.1%) versus those in n‐PDT (55.6%; and 28.4%, respectively). In PDT group, only nine patients deteriorated (6.8%) in comparison with 13 patients in n‐PDT (16.0%). The adverse events (AEs) in PDT group were significantly less than those in n‐PDT group. There were no significant differences in clinical type, gender, age, education level, and family knowledge about WD between the two groups. There were significant differences in AEs and family position toward treatment. Conclusion: Medication Adherence of Chinese WD patients was low. One third of the patients (33.5%) were unable to PDT, and it had an important negative effect on clinical outcome. AEs and family support had an important impact on treatment persistence. [ABSTRACT FROM AUTHOR]

Support Vector Machine (SVM) has always been one of the most successful learning algorithms, with the central idea of maximizing the minimum margin, i.e., the smallest distance from the instances to the classification boundary. However, recent theoretical results disclosed that maximizing the minimum margin does not necessarily lead to better generalization performance, and instead, the margin distribution has been proven to be more crucial. Based on this idea, we propose the Optimal margin Distribution Machine (ODM), which can achieve a better generalization performance by optimizing the margin distribution explicitly. We characterize the margin distribution by the first- and second-order statistics, i.e., the margin mean and variance. The proposed method is a general learning approach which can be applied in any place where SVMs are used, and its superiority is verified both theoretically and empirically in this paper. [ABSTRACT FROM AUTHOR]

Background: Bilateral medial medullary infarction (MMI) is uncommon and bilateral medial pons infarction (MPI) is even rarer. "Heart appearance" on magnetic resonance imaging (MRI) is a characteristic presentation of bilateral medial medullary infarction (MMI).Case Presentation: We present 67-year-old Chinese diabetic and hypertensive female patient affected with "heart appearance-like" infarction in bilateral ponto-medullary junction on MRI. Abnormal signal was observed in the bilateral ponto-medullary junction on T1, T2, fluid-attenuated inversion recovery and apparent diffusion coefficient (ADC). The whole brain digital subtraction angiography (DSA) showed the basilar artery and vertebral artery remained intact. Therefore, we speculated that the bilateral ponto-medullary junction infarction might be caused by the deep perforating branch of the basilar artery.Conclusions: As far as we know, the "heart appearance-like" infraction in bilateral ponto-medullary junction was not reported. Our case also suggests that bilateral ischemic infraction involvement of the medulla and pon is possible even in the context of an intact basilar artery. [ABSTRACT FROM AUTHOR]

Summary of main observation and conclusion: Converting CO2 into value‐added chemicals represents a promising way to alleviate the CO2 derived environmental issues, for which the development of catalysts with high efficiency and recyclability is very desirable. Herein, the catalytic system by combining cobalt source and ionic liquid (IL) has been developed as the efficacious and recyclable catalyst for the carboxylative cyclization of propargylic amine and CO2 to prepare 2‐oxazolinones. In this protocol, various propargylic amines were successfully transformed into the corresponding 2‐oxazolinones with CoBr2 and diethylimidazolium acetate ([EEIM][OAc]) as the catalyst under atmospheric CO2 pressure. It is worth noting that the turnover number (TON) of this transformation can be up to 1740, presumably being attributed to the cooperative effect of the cobalt and IL. Furthermore, the existence of IL enables the catalytic system to be easily recycled to 10 times without losing its activity. [ABSTRACT FROM AUTHOR]

Current deep-learning models are mostly built upon neural networks, i.e. multiple layers of parameterized differentiable non-linear modules that can be trained by backpropagation. In this paper, we explore the possibility of building deep models based on non-differentiable modules such as decision trees. After a discussion about the mystery behind deep neural networks, particularly by contrasting them with shallow neural networks and traditional machine-learning techniques such as decision trees and boosting machines, we conjecture that the success of deep neural networks owes much to three characteristics, i.e. layer-by-layer processing, in-model feature transformation and sufficient model complexity. On one hand, our conjecture may offer inspiration for theoretical understanding of deep learning; on the other hand, to verify the conjecture, we propose an approach that generates deep forest holding these characteristics. This is a decision-tree ensemble approach, with fewer hyper-parameters than deep neural networks, and its model complexity can be automatically determined in a data-dependent way. Experiments show that its performance is quite robust to hyper-parameter settings, such that in most cases, even across different data from different domains, it is able to achieve excellent performance by using the same default setting. This study opens the door to deep learning based on non-differentiable modules without gradient-based adjustment, and exhibits the possibility of constructing deep models without backpropagation. [ABSTRACT FROM AUTHOR]

In many real tasks the features are evolving, with some features vanished and some other features being augmented. For example, in environment monitoring some sensors expired whereas some new ones were deployed; in mobile game recommendation some games dropped whereas some new ones were added. Learning with such incremental and decremental features is crucial but rarely studied, particularly when the data comes like a stream and thus it is infeasible to keep the whole data for optimization. In this paper, we study this challenging problem and present the OPID approach. Our approach attempts to compress important information of vanished features into functions of survived features, and then expand to include the augmented features. It is an one-pass learning approach, which only needs to scan each instance once and does not need to store the whole data, and thus satisfies the evolving streaming data nature. After tackling this problem in one-shot scenario, we then extend it to multi-shot case. Empirical study on a broad range of data sets shows that our approach can address this problem effectively. [ABSTRACT FROM AUTHOR]

Supervised learning techniques construct predictive models by learning from a large number of training examples, where each training example has a label indicating its ground-truth output. Though current techniques have achieved great success, it is noteworthy that in many tasks it is difficult to get strong supervision information like fully ground-truth labels due to the high cost of the data-labeling process. Thus, it is desirable for machine-learning techniques to work with weak supervision. This article reviews some research progress of weakly supervised learning, focusing on three typical types of weak supervision: incomplete supervision, where only a subset of training data is given with labels; inexact supervision, where the training data are given with only coarse-grained labels; and inaccurate supervision, where the given labels are not always ground-truth. [ABSTRACT FROM AUTHOR]

Machine learning is the driving force of the hot artificial intelligence (AI) wave. In an interview with NSR, Prof. Thomas Dietterich, the distinguished professor emeritus of computer science at Oregon State University in the USA, the former president of Association of Advancement of Artificial Intelligence (AAAI, the most prestigious association in the field of artificial intelligence) and the founding president of the International Machine Learning Society, talked about exciting recent advances and technical challenges of machine learning, as well as its big impact on the world. [ABSTRACT FROM AUTHOR]

Performing CO2 conversion in a cost-effective and environmentally benign manner would be promising and remains challenging due to its thermodynamic stability and kinetic inertness. Herein, we would like to summarise significant advances in organic synthesis using CO2 with high catalytic efficiency and excellent selectivity towards the target product mainly during the last five years (2012–2016). Achieving an efficient and selective CO2 conversion depends on the development of metal catalysts (especially functional metal complex catalysis) including main-group metal, typical transition metal and lanthanide series metal as well as organocatalysts e.g. N-heterocyclic carbenes, N-heterocyclic olefins, task-specific ionic liquids, superbases and frustrated Lewis pairs that are able to effectively activate CO2 and/or the substrate on the basis of the mechanistic understanding at the molecular level. This review just covers typical catalytic transformation of CO2, for instance, carboxylation, amidation, hydrogenation, and representative green processes like solvent-less, halogen-free that use CO2 as an ideal carbon-neutral source to prepare valuable compounds with improved atom economy and enhanced sustainability of chemical processes through green catalysis. In particular, in situ catalytic CO2 conversion, i.e. the combination of carbon capture and subsequent conversion, a recent breakthrough in the CO2 chemistry field, is also discussed. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate the prognostic value and pathobiological significance of Glasgow microenvironment score (GMS), a parameter based on tumor stroma percentage and inflammatory cell infiltration, in gastric cancer. Methods: A total of 225 cases of gastric cancer were histologically reviewed, and GMS was evaluated for each case. The association between GMS and patients' survival was investigated. Then the relationship between GMS and mismatch repair (MMR) status, Epstein-Barr virus (EBV) infection were determined using immunohistochemistry (IHC) and in situ hybridization, and the expression of PD1/PD-L1 was examined. Furthermore, the amount of cancer-associated fibroblasts (CAFs), the content and maturity of collagen components were detected using IHC, Picrosirius Red staining and second harmonic generation imaging. Results: GMS was significantly associated with clinical outcomes of gastric cancer, and multivariate analysis indicated that GMS was an independent factor (HR 1.725, P = 0.002). Low GMS was a manifestation of better prognosis and inflammatory tumor microenvironment, which was related to MMR deficiency ( P = 0.042) and EBV infection ( P = 0.032), and within this microenvironment, expression of PD-L1 in carcinoma cells ( P = 0.030) or in inflammatory cells ( P = 0.029) was significantly higher. In contrast, high GMS linked to a poorer survival and desmoplastic stroma, in which there existed markedly increased CAFs and collagen deposition. Conclusion: GMS can serve as a useful prognostic factor for gastric cancer, and according to GMS, the tumor microenvironment in this cancer type may be partially classified as inflammatory or desmoplastic microenvironment that possesses different pathobiological features. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is an insidious onset neurodegenerative disease affecting approximately 1% of the population over the age of 65. So far available therapies for PD have only aimed at improving or alleviating symptoms, but not at slowing, preventing, and reversing the course of PD. Recently, some studies have indicated that the levels and activation of Abelson non-receptor tyrosine kinase (c-Abl, Abl1) were up-regulated in the brain tissue of patients with PD and demonstrated that c-Abl inhibitors could improve motor behavior, prevent the loss of dopamine neurons, inhibit phosphorylation of Cdk5, regulate α-synuclein phosphorylation and clearance, inhibit the tyrosine phosphorylation of parkin and decrease parkin substrate, for example, PARIS (zinc finger protein 746), AIMP2 (aminoacyl-tRNA synthetase-interacting multifunctional protein type2), FBP1 (fuse-binding protein 1), and synphilin-1. Therefore, we review the mechanism of the c-Abl inhibitor in PD and conclude that c-Abl inhibitors may be a potential treatment in PD and other neurodegenerative disease. [ABSTRACT FROM AUTHOR]

A silver(I)-catalyzed three-component reaction of propargylic alcohols, CO2, and monohydric alcohols was successfully developed for the synthesis of β-oxopropyl carbonates. As such, a series of β-oxopropyl carbonates were exclusively produced in excellent yields (up to 98 %), even under atmospheric pressure of CO2. The silver catalyst works efficiently for both the carboxylative cyclization of propargylic alcohols with CO2 and subsequent transesterification of α-alkylidene cyclic carbonates with monohydric alcohols; thus this tandem process performs smoothly under mild conditions. This work provides a versatile and thermodynamically favorable approach to dissymmetric dialkyl carbonates. [ABSTRACT FROM AUTHOR]

In many real-world applications, different types of misclassification usually suffer from different costs, but the accurate cost is often hard to be determined and usually one can only get an interval-estimation like that one type of mistake is about 5 to 10 times more serious than the other type. On the other hand, there are usually abundant unlabeled data available, leading to great research effort about semi-supervised learning. It is noticeable that cost interval and unlabeled data usually appear simultaneously in practice tasks; however, there is rare study tackling them together. In this paper, we propose the cisLDM approach which is able to handle cost interval and exploit unlabeled data in a principled way. Rather than maximizing the minimum margin like traditional large margin classifiers, cisLDM tries to optimize the margin distribution on both labeled and unlabeled data when minimizing the worst-case total-cost and the mean total-cost simultaneously according to the cost interval. Experiments on a broad range of datasets and cost settings exhibit the impressive performance of cisLDM. In particular, cisLDM is able to reduce 47 percent more total-cost than standard SVM and 27 percent more total-cost than cost-sensitive semi-supervised SVM which assumes the true cost value is known in advance. [ABSTRACT FROM PUBLISHER]

A new nonlinear beam-column element capable of considering the shear deformation is proposed under the concept of consistent deformation. For the traditional displacement interpolation function, the beam-column element produces membrane locking under large deformation and shear locking when the element becomes slender. To eliminate the membrane and shear locking, force equilibrium equations are employed to derive the displacement function. Numerical examples herein show that membrane locking in the traditional nonlinear beam-column element could cause a considerable error. Comparison of the present improved formulae based on the Timoshenko beam theory with that based on the Euler-Bernoulli beam theory indicates that the present approach requires several additional parameters to consider shear deformation and it is suitable for computer analysis readily for incorporation into the frames analysis software using the co-rotational approach for large translations and rotations. The examples confirm that the proposed element has higher accuracy and numerical efficiency. [ABSTRACT FROM AUTHOR]

To reevaluate the association between the costimulatory molecule cytotoxic T lymphocyte-associated antigen4 ( CTLA4) single nucleotide polymorphism (SNP) +49A/G and acute rejection (AR) in renal transplantation, nine studies published before June 2013 were analyzed. Meta-analysis and cumulative meta-analysis (metacum) were performed for each genotype in a random/fixed effect model. The combined odds ratios (OR) with 95% confidence intervals (CI) were calculated to estimate the strength of the association. In the sensitivity analysis, a single study involved in the meta-analysis was deleted each time to investigate the influence of the individual data sets on the pooled ORs. Meta-analysis regression was used for some influence factors, such as year of publication, total number in each group (AR group and control group), ethnicity, the ratio of GG to GA + AA, the ratio of G to A in CTLA4 +49A/G. Overall, a significant correlation was noted between the CTLA4 SNP (+49A/G) and the risk of AR (for GG vs. AG + AA: OR = 1.35, 95% CI = 1.05-1.73, p = 0.02; for G vs. A: OR = 1.21, 95% CI = 1.03-1.42, p = 0.02), especially in the Asian subgroup (for GG vs. AG + AA: OR = 1.79, 95% CI = 1.15-2.78, p = 0.009; for G vs. A: OR = 1.47, 95% CI = 1.04-2.07, p = 0.03). Of the influence factors, the ratio of GG to GA+AA ( p = 0.046) and the ratio of G to A ( p = 0.017) were significant factors. In conclusion, our results suggest that CTLA4 +49A/G contribute to the risk of AR following renal transplantation. [ABSTRACT FROM AUTHOR]

It is usually expected that learning performance can be improved by exploiting unlabeled data, particularly when the number of labeled data is limited. However, it has been reported that, in some cases existing semi-supervised learning approaches perform even worse than supervised ones which only use labeled data. For this reason, it is desirable to develop safe semi-supervised learning approaches that will not significantly reduce learning performance when unlabeled data are used. This paper focuses on improving the safeness of semi-supervised support vector machines (S3VMs). First, the S3VM-us approach is proposed. It employs a conservative strategy and uses only the unlabeled instances that are very likely to be helpful, while avoiding the use of highly risky ones. This approach improves safeness but its performance improvement using unlabeled data is often much smaller than S3VMs. In order to develop a safe and well-performing approach, we examine the fundamental assumption of S3VMs, i.e., low-density separation. Based on the observation that multiple good candidate low-density separators may be identified from training data, safe semi-supervised support vector machines (S4VMs) are here proposed. This approach uses multiple low-density separators to approximate the ground-truth decision boundary and maximizes the improvement in performance of inductive SVMs for any candidate separator. Under the assumption employed by S3VMs, it is here shown that S4VMs are provably safe and that the performance improvement using unlabeled data can be maximized. An out-of-sample extension of S4VMs is also presented. This extension allows S4VMs to make predictions on unseen instances. Our empirical study on a broad range of data shows that the overall performance of S4VMs is highly competitive with S3VMs, whereas in contrast to S3VMs which hurt performance significantly in many cases, S4VMs rarely perform worse than inductive SVMs. [ABSTRACT FROM AUTHOR]

Multi-label learning studies the problem where each example is represented by a single instance while associated with a set of labels simultaneously. During the past decade, significant amount of progresses have been made toward this emerging machine learning paradigm. This paper aims to provide a timely review on this area with emphasis on state-of-the-art multi-label learning algorithms. Firstly, fundamentals on multi-label learning including formal definition and evaluation metrics are given. Secondly and primarily, eight representative multi-label learning algorithms are scrutinized under common notations with relevant analyses and discussions. Thirdly, several related learning settings are briefly summarized. As a conclusion, online resources and open research problems on multi-label learning are outlined for reference purposes. [ABSTRACT FROM PUBLISHER]

Abstract: Margin theory provides one of the most popular explanations to the success of AdaBoost, where the central point lies in the recognition that margin is the key for characterizing the performance of AdaBoost. This theory has been very influential, e.g., it has been used to argue that AdaBoost usually does not overfit since it tends to enlarge the margin even after the training error reaches zero. Previously the minimum margin bound was established for AdaBoost, however, Breiman (1999) [9] pointed out that maximizing the minimum margin does not necessarily lead to a better generalization. Later, Reyzin and Schapire (2006) [37] emphasized that the margin distribution rather than minimum margin is crucial to the performance of AdaBoost. In this paper, we first present the kth margin bound and further study on its relationship to previous work such as the minimum margin bound and Emargin bound. Then, we improve the previous empirical Bernstein bounds (Audibert et al. 2009; Maurer and Pontil, 2009) [2,30], and based on such findings, we defend the margin-based explanation against Breimanʼs doubts by proving a new generalization error bound that considers exactly the same factors as Schapire et al. (1998) [39] but is sharper than Breimanʼs (1999) [9] minimum margin bound. By incorporating factors such as average margin and variance, we present a generalization error bound that is heavily related to the whole margin distribution. We also provide margin distribution bounds for generalization error of voting classifiers in finite VC-dimension space. [Copyright &y& Elsevier]

Abstract: Multi-label learning has attracted much attention during the past few years. Many multi-label approaches have been developed, mostly working with surrogate loss functions because multi-label loss functions are usually difficult to optimize directly owing to their non-convexity and discontinuity. These approaches are effective empirically, however, little effort has been devoted to the understanding of their consistency, i.e., the convergence of the risk of learned functions to the Bayes risk. In this paper, we present a theoretical analysis on this important issue. We first prove a necessary and sufficient condition for the consistency of multi-label learning based on surrogate loss functions. Then, we study the consistency of two well-known multi-label loss functions, i.e., ranking loss and hamming loss. For ranking loss, our results disclose that, surprisingly, none of convex surrogate loss is consistent; we present the partial ranking loss, with which some surrogate losses are proven to be consistent. We also discuss on the consistency of univariate surrogate losses. For hamming loss, we show that two multi-label learning methods, i.e., one-vs-all and pairwise comparison, which can be regarded as direct extensions from multi-class learning, are inconsistent in general cases yet consistent under the dominating setting, and similar results also hold for some recent multi-label approaches that are variations of one-vs-all. In addition, we discuss on the consistency of learning approaches that address multi-label learning by decomposing into a set of binary classification problems. [Copyright &y& Elsevier]

Abstract: In this paper, we propose the MIML (Multi-Instance Multi-Label learning) framework where an example is described by multiple instances and associated with multiple class labels. Compared to traditional learning frameworks, the MIML framework is more convenient and natural for representing complicated objects which have multiple semantic meanings. To learn from MIML examples, we propose the MimlBoost and MimlSvm algorithms based on a simple degeneration strategy, and experiments show that solving problems involving complicated objects with multiple semantic meanings in the MIML framework can lead to good performance. Considering that the degeneration process may lose information, we propose the D-MimlSvm algorithm which tackles MIML problems directly in a regularization framework. Moreover, we show that even when we do not have access to the real objects and thus cannot capture more information from real objects by using the MIML representation, MIML is still useful. We propose the InsDif and SubCod algorithms. InsDif works by transforming single-instances into the MIML representation for learning, while SubCod works by transforming single-label examples into the MIML representation for learning. Experiments show that in some tasks they are able to achieve better performance than learning the single-instances or single-label examples directly. [Copyright &y& Elsevier]

Co-training is one of the major semi-supervised learning paradigms that iteratively trains two classifiers on two different views, and uses the predictions of either classifier on the unlabeled examples to augment the training set of the other. During the co-training process, especially in initial rounds when the classifiers have only mediocre accuracy, it is quite possible that one classifier will receive labels on unlabeled examples erroneously predicted by the other classifier. Therefore, the performance of co-training style algorithms is usually unstable. In this paper, the problem of how to reliably communicate labeling information between different views is addressed by a novel co-training algorithm named CoTrade. In each labeling round, CoTrade carries out the label communication process in two steps. First, confidence of either classifier's predictions on unlabeled examples is explicitly estimated based on specific data editing techniques. Secondly, a number of predicted labels with higher confidence of either classifier are passed to the other one, where certain constraints are imposed to avoid introducing undesirable classification noise. Experiments on several real-world datasets across three domains show that CoTrade can effectively exploit unlabeled data to achieve better generalization performance. [ABSTRACT FROM AUTHOR]

Abstract: Enrichment of cancer stem cells for studies of carcinogenesis remains a difficult issue. We hypothesized that the unique features of cancer stem cells (CSCs) may allow formation of their colonies in vitro with distinct morphology. We therefore investigated the possibility to use morphological diversity of colonies to identify and enrich CSCs from cultured malignant human glioma cells. We found that a small proportion of the cells from a human glioma cell line U251 formed tight and round-shaped colonies in culture. Most cells in such colonies were capable of self-renewal, generating tumor spheres and differentiating into lineages with markers for neurons, astrocytes and oligodendrocytes. In addition, several neural stem cell-related genes were highly expressed by tumor cells in those tight colonies. Our results thus demonstrate a novel approach to the identification and enrichment of CSCs based on unique morphology of their colonies formed in vitro. [Copyright &y& Elsevier]

Abstract: Evolutionary algorithms (EA) have been shown to be very effective in solving practical problems, yet many important theoretical issues of them are not clear. The expected first hitting time is one of the most important theoretical issues of evolutionary algorithms, since it implies the average computational time complexity. In this paper, we establish a bridge between the expected first hitting time and another important theoretical issue, i.e., convergence rate. Through this bridge, we propose a new general approach to estimating the expected first hitting time. Using this approach, we analyze EAs with different configurations, including three mutation operators, with/without population, a recombination operator and a time variant mutation operator, on a hard problem. The results show that the proposed approach is helpful for analyzing a broad range of evolutionary algorithms. Moreover, we give an explanation of what makes a problem hard to EAs, and based on the recognition, we prove the hardness of a general problem. [Copyright &y& Elsevier]

The influence of non-isothermal melt crystallization on thermal behavior and isothermal melt crystallization kinetics of poly-L-lactide (PLLA) were investigated by differential scanning calorimetry (DSC), polarizing micrograph (POM) and x-ray diffraction (XRD). Crystallization performed at lower cooling rates (2°C·min-1) is accompanied by a variation of the kinetics around 118°C. The glass transition temperature of PLLA decreases with increase of cooling rate, and the crystallinity at the end of crystallization increases with decreasing cooling rate. The size of PLLA spherulites increases with a decrease in the cooling rate, and PLLA becomes almost amorphous cooled at rapid rate (>10°C·min-1). PLLA exhibits an Avrami crystallization exponent n = 3.01±0.13 in isothermal crystallization in the range from 90°C to 140°C. According to Hoffman-Lauritzen theory, two crystallization regime are identified with a transition temperature occurring at 118°C, and the value of Kg(II)/Kg(III) is 2.17 [Kg(II) = 6.025 × 105K2, Kg(III) = 1.307 × 106 K2]. [ABSTRACT FROM AUTHOR]

Abstract: Multi-label learning originated from the investigation of text categorization problem, where each document may belong to several predefined topics simultaneously. In multi-label learning, the training set is composed of instances each associated with a set of labels, and the task is to predict the label sets of unseen instances through analyzing training instances with known label sets. In this paper, a multi-label lazy learning approach named ML-KNN is presented, which is derived from the traditional K-nearest neighbor (KNN) algorithm. In detail, for each unseen instance, its K nearest neighbors in the training set are firstly identified. After that, based on statistical information gained from the label sets of these neighboring instances, i.e. the number of neighboring instances belonging to each possible class, maximum a posteriori (MAP) principle is utilized to determine the label set for the unseen instance. Experiments on three different real-world multi-label learning problems, i.e. Yeast gene functional analysis, natural scene classification and automatic web page categorization, show that ML-KNN achieves superior performance to some well-established multi-label learning algorithms. [Copyright &y& Elsevier]

Abstract: Ensemble methods that train multiple learners and then combine their predictions have been shown to be very effective in supervised learning. This paper explores ensemble methods for unsupervised learning. Here, an ensemble comprises multiple clusterers, each of which is trained by k-means algorithm with different initial points. The clusters discovered by different clusterers are aligned, i.e. similar clusters are assigned with the same label, by counting their overlapped data items. Then, four methods are developed to combine the aligned clusterers. Experiments show that clustering performance could be significantly improved by ensemble methods, where utilizing mutual information to select a subset of clusterers for weighted voting is a nice choice. Since the proposed methods work by analyzing the clustering results instead of the internal mechanisms of the component clusterers, they are applicable to diverse kinds of clustering algorithms. [Copyright &y& Elsevier]

Abstract: Recently, a new technique called two-dimensional principal component analysis (2DPCA) was proposed for face representation and recognition. The main idea behind 2DPCA is that it is based on 2D matrices as opposed to the standard PCA, which is based on 1D vectors. Although 2DPCA obtains higher recognition accuracy than PCA, a vital unresolved problem of 2DPCA is that it needs many more coefficients for image representation than PCA. In this paper, we first indicate that 2DPCA is essentially working in the row direction of images, and then propose an alternative 2DPCA which is working in the column direction of images. By simultaneously considering the row and column directions, we develop the two-directional 2DPCA, i.e. , for efficient face representation and recognition. Experimental results on ORL and a subset of FERET face databases show that achieves the same or even higher recognition accuracy than 2DPCA, while the former needs a much reduced coefficient set for image representation than the latter. [Copyright &y& Elsevier]

In this paper, the generalized projection function (GPF) is defined. Both the integral projection function (IPF) and the variance projection function (VPF) can be viewed as special cases of GPF. Another special case of GPF, i.e. the hybrid projection function (HPF), is developed through experimentally determining the optimal parameters of GPF. Experiments on three face databases show that IPF, VPF, and HPF are all effective in eye detection. Nevertheless, HPF is better than VPF, while VPF is better than IPF. Moreover, IPF is found to be more effective on occidental than on oriental faces, and VPF is more effective on oriental than on occidental faces. Analysis of the detections shows that this effect may be owed to the shadow of the noses and eyeholes of different races of people. [Copyright &y& Elsevier]

The ultimate load of a typical steel frame is dependent on the geometrically nonlinear and material yielding effects. The complexity for considering material yielding by the plastic hinge approach is the unknown location of the plastic hinge, which can occur at the ends or any position along the element length. For the latter case, a member is divided into many elements in order to approximate the location of a plastic hinge. This process is tedious, inconvenient to use, and involves extensive computer time. Further, the strength check for sectional capacity using the LRFD code requires an assumption of the K factor or the effective length ratio, which further complicates a computer analysis for the ultimate load of a steel frame. To describe the formation of a plastic hinge along an element in a member at the ultimate limit state, a single element capable of modeling the P-δ effect as well as the formation of the plastic hinge is needed. This paper adopts a simple concept of superimposition of triangular deflected shapes due to the formation of plastic hinge to the fifth order deflection shape for elastic deflection to yield the final deflection of the element, the plastic pointwise equilibrium polynomial (PPEP) element. Equilibrium of moment and shear at midspan of an element is maintained for accurate modeling of the P-δ effect in the tangent and secant stiffness. The robustness, accuracy and reliability of the developed element are demonstrated in a number of worked examples. [ABSTRACT FROM AUTHOR]

An isolated member capacity check for beam columns may not correctly reflect the true behavior of a structure undergoing large deflection and material yielding. This paper extends the previous work by the authors on geometrically nonlinear analysis of skeletal structures to combined geometrically and material nonlinear analysis of slender frames using a single element per member. In the proposed element, three plastic hinges are allowed to form in an element with two at the two ends and one at the location of maximum combined stress due to axial force and moment. For calibration against the load capacity of a single member in the load and resistance factor design and allowable stress design standards, the present results perform excellently while, in the case of ultimate and collapse analysis of steel frames, the present method gives accurate results allowing for interaction between all members with geometric and material nonlinearities. The formulation is capable of conducting an elastoplastic buckling analysis of a beam column modeled by one element per member, which is not available in literature. [ABSTRACT FROM AUTHOR]

In this paper, genetic algorithm is used to help improve the tolerance of feedforward neural networks against an open .fault. The proposed method does not explicitly add any redundancy to the network, nor does it modify the training algorithm. Experiments show that it may profit the fault tolerance as well as the generalisation ability of neural networks. [ABSTRACT FROM AUTHOR]

In this paper an efficient face candidates selector is proposed for face detection tasks in still gray level images. The proposed method acts as a selective attentional mechanism. Eye-analogue segments at a given scale are discovered by finding regions which are roughly as large as real eyes and are darker than their neighborhoods. Then a pair of eye-analogue segments are hypothesized to be eyes in a face and combined into a face candidate if their placement is consistent with the anthropological characteristic of human eyes. The proposed method is robust in that it can deal with illumination changes and moderate rotations. A subset of the FERET data set and the BioID face database are used to evaluate the proposed method. The proposed face candidates selector is successful in 98.75% and 98.6% cases, respectively. [Copyright &y& Elsevier]

Neural network ensemble can significantly improve the generalization ability of neural network based systems. However, its comprehensibility is even worse than that of a single neural network because it comprises a collection of individual neural networks. In this paper, an approach named REFNE is proposed to improve the comprehensibility of trained neural network ensembles that perform classification tasks. REFNE utilizes the trained ensembles to generate instances and then extracts symbolic rules from those instances. It gracefully breaks the ties made by individual neural networks in prediction. It also employs specific discretization scheme, rule form, and fidelity evaluation mechanism. Experiments show that with different configurations, REFNE can extract rules with good fidelity that well explain the function of trained neural network ensembles, or rules with strong generalization ability that are even better than the trained neural network ensembles in prediction. [ABSTRACT FROM AUTHOR]

This paper reviews three recent books on data mining written from three different perspectives, i.e., databases, machine learning, and statistics. Although the exploration in this paper is suggestive instead of conclusive, it reveals that besides some common properties, different perspectives lay strong emphases on different aspects of data mining. The emphasis of the database perspective is on efficiency because this perspective strongly concerns the whole discovery process and huge data volume. The emphasis of the machine learning perspective is on effectiveness because this perspective is heavily attracted by substantive heuristics working well in data analysis although they may not always be useful. As for the statistics perspective, its emphasis is on validity because this perspective cares much for mathematical soundness behind mining methods. [Copyright &y& Elsevier]

At present there are many methods that could deal well with frontal view face recognition. However, most of them cannot work well when there is only one training image per person. In this paper, an extension of the eigenface technique, i.e. projection-combined principal component analysis, (PC)2A, is proposed. (PC)2A combines the original face image with its horizontal and vertical projections and then performs principal component analysis on the enriched version of the image. It requires less computational cost than the standard eigenface technique and experimental results show that on a gray-level frontal view face database where each person has only one training image, (PC)2A achieves 3–5% higher accuracy than the standard eigenface technique through using 10–15% fewer eigenfaces. [Copyright &y& Elsevier]

In this paper, a hybrid learning approach named hybrid decision tree (HDT) is proposed. HDT simulates human reasoning by using symbolic learning to do qualitative analysis and using neural learning to do subsequent quantitative analysis. It generates the trunk of a binary HDT according to the binary information gain ratio criterion in an instance space defined by only original unordered attributes. If unordered attributes cannot further distinguish training examples falling into a leaf node whose diversity is beyond the diversity-threshold, then the node is marked as a dummy node. After all those dummy nodes are marked, a specific feedforward neural network named Fannc that is trained in an instance space defined by only original ordered attributes is exploited to accomplish the learning task. Moreover, this paper distinguishes three kinds of incremental learning tasks. Two incremental learning procedures designed for example-incremental learning with different storage requirements are provided, which enables HDT to deal gracefully with data sets where new data are frequently appended. Also a hypothesis-driven constructive induction mechanism is provided, which enables HDT to generate compact concept descriptions. [Copyright &y& Elsevier]

Neural network ensemble is a learning paradigm where many neural networks are jointly used to solve a problem. In this paper, the relationship between the ensemble and its component neural networks is analyzed from the context of both regression and classification, which reveals that it may be better to ensemble many instead of all of the neural networks at hand. This result is interesting because at present, most approaches ensemble all the available neural networks for prediction. Then, in order to show that the appropriate neural networks for composing an ensemble can be effectively selected from a set of available neural networks, an approach named GASEN is presented. GASEN trains a number of neural networks at first. Then it assigns random weights to those networks and employs genetic algorithm to evolve the weights so that they can characterize to some extent the fitness of the neural networks in constituting an ensemble. Finally it selects some neural networks based on the evolved weights to make up the ensemble. A large empirical study shows that, compared with some popular ensemble approaches such as Bagging and Boosting, GASEN can generate neural network ensembles with far smaller sizes but stronger generalization ability. Furthermore, in order to understand the working mechanism of GASEN, the bias-variance decomposition of the error is provided in this paper, which shows that the success of GASEN may lie in that it can significantly reduce the bias as well as the variance. [Copyright &y& Elsevier]

Neural network ensemble is a learning paradigm where a collection of neural networks is trained for the same task. In this paper, the relationship between the generalization ability of the neural network ensemble and the correlation of the individual neural networks constituting the ensemble is analyzed in the context of combining neural regression estimators, which reveals that ensembling a selective subset of trained networks is superior to ensembling all the trained networks in some cases. Based on such recognition, an approach named GASEN is proposed. GASEN trains a number of individual neural networks at first. Then it assigns random weights to the individual networks and employs a genetic algorithm to evolve those weights so that they can characterize to some extent the importance of the individual networks in constituting an ensemble. Finally it selects an optimum subset of individual networks based on the evolved weights to make up the ensemble. Experimental results show that, comparing with a popular ensemble approach, i.e. averaging all, and a theoretically optimum selective ensemble approach, i.e. enumerating, GASEN has preferable performance in generating ensembles with strong generalization ability in relatively small computational cost. This paper also analyzes the working mechanism of GASEN from the view of error-ambiguity decomposition, which reveals that GASEN improves generalization ability mainly through reducing the average generalization error of the individual neural networks constituting the ensemble. [ABSTRACT FROM AUTHOR]

Complex-valued neural networks have attracted increasing attention in recent years, while it remains open on the advantages of complex-valued neural networks in comparison with real-valued networks. This work takes one step on this direction by introducing the complex-reaction network with fully-connected feed-forward architecture. We prove the universal approximation property for complex-reaction networks, and show that a class of radial functions can be approximated by a complex-reaction network using the polynomial number of parameters, whereas real-valued networks need at least exponential parameters to reach the same approximation level. For empirical risk minimization, we study the landscape and convergence of complex gradient descents. Our theoretical result shows that the critical point set of complex-reaction networks is a proper subset of that of real-valued networks, which may show some insights on finding the optimal solutions more easily for complex-reaction networks. [ABSTRACT FROM AUTHOR]

Recent research has used margin theory to analyze the generalization performance for deep neural networks (DNNs). The existed results are almost based on the spectrally-normalized minimum margin. However, optimizing the minimum margin ignores a mass of information about the entire margin distribution, which is crucial to generalization performance. In this paper, we prove a generalization upper bound dominated by the statistics of the entire margin distribution. Compared with the minimum margin bounds, our bound highlights an important measure for controlling the complexity, which is the ratio of the margin standard deviation to the expected margin. We utilize a convex margin distribution loss function on the deep neural networks to validate our theoretical results by optimizing the margin ratio. Experiments and visualizations confirm the effectiveness of our approach and the correlation between generalization gap and margin ratio. [ABSTRACT FROM AUTHOR]

An accurate pointwise equilibrating polynomial element with springs connected to its ends to model semirigid connections is derived. The formulated element and coded computer program for second-order nonlinear analysis of flexibly jointed steel frames is employed to study a variety of problems using one element per member, which is not possible by the conventional cubic Hermitian function and most of the currently available elements. When incorporated into a nonlinear-analysis computer program, the versatility of the program is greatly improved as the convergence rate is increased, and the modeling effort for accurate solution is reduced considerably because the element stiffness remains accurate under high axial load. The capability of the proposed element in handling a harsh benchmark problem is also verified in one of the examples. [ABSTRACT FROM AUTHOR]

In practical stability design and analysis of steel members and structures, one must allow for member imperfection. Various national design codes impose different values of initial imperfection for member-strength determination, such as 0.001 of the member length in the 1986 Load and Resistance Factor Design Specification for Structural Steel Buildings . This paper presents a new method of including the effects of initial imperfection in the element stiffness without needing to adopt a curved-element formulation, which is deficient for members under high axial load, or to divide a member into two or more straight elements in order to simulate member imperfection. A very considerable savings and convenience in data-manipulation effort and computer time can be achieved when using the proposed element. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a novel subspace method called diagonal principal component analysis (DiaPCA) is proposed for face recognition. In contrast to standard PCA, DiaPCA directly seeks the optimal projective vectors from diagonal face images without image-to-vector transformation. While in contrast to 2DPCA, DiaPCA reserves the correlations between variations of rows and those of columns of images. Experiments show that DiaPCA is much more accurate than both PCA and 2DPCA. Furthermore, it is shown that the accuracy can be further improved by combining DiaPCA with 2DPCA. [Copyright &y& Elsevier]

Learning with streaming data has attracted much attention during the past few years. Though most studies consider data stream with fixed features, in real practice the features may be evolvable. For example, features of data gathered by limited-lifespan sensors will change when these sensors are substituted by new ones. In this article, we propose a novel learning paradigm: Feature Evolvable Streaming Learning where old features would vanish and new features would occur. Rather than relying on only the current features, we attempt to recover the vanished features and exploit it to improve performance. Specifically, we learn a mapping from the overlapping period to recover old features and then we learn two models from the recovered features and the current features, respectively. To benefit from the recovered features, we develop two ensemble methods. In the first method, we combine the predictions from two models and theoretically show that with the assistance of old features, the performance on new features can be improved and we provide a tighter bound when the loss function is exponentially concave. In the second approach, we dynamically select the best single prediction and establish a better performance guarantee when the best model switches. Experiments on both synthetic and real data validate the effectiveness of our proposal. [ABSTRACT FROM AUTHOR]