Search

A growing number of applications, e.g., video surveillance and medical image analysis, require training recognition systems from large amounts of weakly annotated data, while some targeted interactions with a domain expert are allowed to improve the training process. In such cases, active learning (AL) can reduce labeling costs for training a classifier by querying the expert to provide the labels of most informative instances. This paper focuses on AL methods for instance classification problems in multiple instance learning (MIL), where data are arranged into sets, called bags, which are weakly labeled. Most AL methods focus on single-instance learning problems. These methods are not suitable for MIL problems because they cannot account for the bag structure of data. In this paper, new methods for bag-level aggregation of instance informativeness are proposed for multiple instance AL (MIAL). The aggregated informativeness method identifies the most informative instances based on classifier uncertainty and queries bags incorporating the most information. The other proposed method, called cluster-based aggregative sampling, clusters data hierarchically in the instance space. The informativeness of instances is assessed by considering bag labels, inferred instance labels, and the proportion of labels that remain to be discovered in clusters. Both proposed methods significantly outperform reference methods in extensive experiments using benchmark data from several application domains. Results indicate that using an appropriate strategy to address MIAL problems yields a significant reduction in the number of queries needed to achieve the same level of performance as single-instance AL methods. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a new classification algorithm called the optimization of distribution differences (ODD). The algorithm aims to find a transformation from the feature space to a new space where the instances in the same class are as close as possible to one another, whereas the gravity centers of these classes are as far as possible from one another. This aim is formulated as a multiobjective optimization problem that is solved by a hybrid of an evolutionary strategy and the quasi-Newton method. The choice of the transformation function is flexible and could be any continuous space function. We experiment with a linear and a nonlinear transformation in this paper. We show that the algorithm can outperform eight other classification methods, namely naive Bayes, support vector machines, linear discriminant analysis, multilayer perceptrons, decision trees, and $k$ -nearest neighbors, and two recently proposed classification methods, in 12 standard classification data sets. Our results show that the method is less sensitive to the imbalanced number of instances compared with these methods. We also show that ODD maintains its performance better than other classification methods in these data sets and hence offers a better generalization ability. [ABSTRACT FROM AUTHOR]

The class imbalance issue has been a persistent problem in machine learning that hinders the accurate predictive analysis of data in many real-world applications. The class imbalance problem exists when the number of instances present in a class (or classes) is significantly fewer than the number of instances belonging to another class (or classes). Sufficiently recognizing the minority class during classification is a problem as most algorithms employed to learn from data input are biased toward the majority class. The underlying issue is made more complex with the presence of data difficult factors embedded in such data input. This paper presents a novel and effective ensemble-based method for dealing with the class imbalance problem. This paper is motivated by the idea of moving the oversampling from the data level to the algorithm level, instead of increasing the minority instances in the data sets, the algorithms in this paper aims to “oversample the classification ensemble” by increasing the number of classifiers that represent the minority class in the ensemble, i.e., random forest. The proposed biased random forest algorithm employs the nearest neighbor algorithm to identify the critical areas in a given data set. The standard random forest is then fed with more random trees generated based on the critical areas. The results show that the proposed algorithm is very effective in dealing with the class imbalance problem. [ABSTRACT FROM AUTHOR]

In recent years, the performance of semisupervised learning (SSL) has been theoretically investigated. However, most of this theoretical development has focused on binary classification problems. In this paper, we take it a step further by extending the work of Castelli and Cover to the multiclass paradigm. In particular, we consider the key problem in SSL of classifying an unseen instance x into one of $K$ different classes, using a training data set sampled from a mixture density distribution and composed of $l$ labeled records and $u$ unlabeled examples. Even under the assumption of identifiability of the mixture and having infinite unlabeled examples, labeled records are needed to determine the $K$ decision regions. Therefore, in this paper, we first investigate the minimum number of labeled examples needed to accomplish that task. Then, we propose an optimal multiclass learning algorithm, which is a generalization of the optimal procedure proposed in the literature for binary problems. Finally, we make use of this generalization to study the probability of error when the binary class constraint is relaxed. [ABSTRACT FROM PUBLISHER]

As each imbalanced classification problem comes with its own set of challenges, the measure used to evaluate classifiers must be individually selected. To help researchers make this decision in an informed manner, experimental and theoretical investigations compare general properties of measures. However, existing studies do not analyze changes in measure behavior imposed by different imbalance ratios. Moreover, several characteristics of imbalanced data streams, such as the effect of dynamically changing class proportions, have not been thoroughly investigated from the perspective of different metrics. In this paper, we study measure dynamics by analyzing changes of measure values, distributions, and gradients with diverging class proportions. For this purpose, we visualize measure probability mass functions and gradients. In addition, we put forward a histogram-based normalization method that provides a unified, probabilistic interpretation of any measure over data sets with different class distributions. The results of analyzing eight popular classification measures show that the effect class proportions have on each measure is different and should be taken into account when evaluating classifiers. Apart from highlighting imbalance-related properties of each measure, our study shows a direct connection between class ratio changes and certain types of concept drift, which could be influential in designing new types of classifiers and drift detectors for imbalanced data streams. [ABSTRACT FROM AUTHOR]

This paper proposes a novel end-to-end learning model, called skip-connected covariance (SCCov) network, for remote sensing scene classification (RSSC). The innovative contribution of this paper is to embed two novel modules into the traditional convolutional neural network (CNN) model, i.e., skip connections and covariance pooling. The advantages of newly developed SCCov are twofold. First, by means of the skip connections, the multi-resolution feature maps produced by the CNN are combined together, which provides important benefits to address the presence of large-scale variance in RSSC data sets. Second, by using covariance pooling, we can fully exploit the second-order information contained in such multi-resolution feature maps. This allows the CNN to achieve more representative feature learning when dealing with RSSC problems. Experimental results, conducted using three large-scale benchmark data sets, demonstrate that our newly proposed SCCov network exhibits very competitive or superior classification performance when compared with the current state-of-the-art RSSC techniques, using a much lower amount of parameters. Specifically, our SCCov only needs 10% of the parameters used by its counterparts. [ABSTRACT FROM AUTHOR]

The most popular tools for stream data mining are based on decision trees. In previous 15 years, all designed methods, headed by the very fast decision tree algorithm, relayed on Hoeffding’s inequality and hundreds of researchers followed this scheme. Recently, we have demonstrated that although the Hoeffding decision trees are an effective tool for dealing with stream data, they are a purely heuristic procedure; for example, classical decision trees such as ID3 or CART cannot be adopted to data stream mining using Hoeffding’s inequality. Therefore, there is an urgent need to develop new algorithms, which are both mathematically justified and characterized by good performance. In this paper, we address this problem by developing a family of new splitting criteria for classification in stationary data streams and investigating their probabilistic properties. The new criteria, derived using appropriate statistical tools, are based on the misclassification error and the Gini index impurity measures. The general division of splitting criteria into two types is proposed. Attributes chosen based on type- $I$ splitting criteria guarantee, with high probability, the highest expected value of split measure. Type- $II$ criteria ensure that the chosen attribute is the same, with high probability, as it would be chosen based on the whole infinite data stream. Moreover, in this paper, two hybrid splitting criteria are proposed, which are the combinations of single criteria based on the misclassification error and Gini index. [ABSTRACT FROM AUTHOR]

Multiview learning has shown promising potential in many applications. However, most techniques are focused on either view consistency, or view diversity. In this paper, we introduce a novel multiview boosting algorithm, called Boost.SH, that computes weak classifiers independently of each view but uses a shared weight distribution to propagate information among the multiple views to ensure consistency. To encourage diversity, we introduce randomized Boost.SH and show its convergence to the greedy Boost.SH solution in the sense of minimizing regret using the framework of adversarial multiarmed bandits. We also introduce a variant of Boost.SH that combines decisions from multiple experts for recommending views for classification. We propose an expert strategy for multiview learning based on inverse variance, which explores both consistency and diversity. Experiments on biometric recognition, document categorization, multilingual text, and yeast genomic multiview data sets demonstrate the advantage of Boost.SH (85%) compared with other boosting algorithms like AdaBoost (82%) using concatenated views and substantially better than a multiview kernel learning algorithm (74%). [ABSTRACT FROM PUBLISHER]

Random forests (RFs) are recognized as one type of ensemble learning method and are effective for the most classification and regression tasks. Despite their impressive empirical performance, the theory of RFs has yet been fully proved. Several theoretically guaranteed RF variants have been presented, but their poor practical performance has been criticized. In this paper, a novel RF framework is proposed, named Bernoulli RFs (BRFs), with the aim of solving the RF dilemma between theoretical consistency and empirical performance. BRF uses two independent Bernoulli distributions to simplify the tree construction, in contrast to the RFs proposed by Breiman. The two Bernoulli distributions are separately used to control the splitting feature and splitting point selection processes of tree construction. Consequently, theoretical consistency is ensured in BRF, i.e., the convergence of learning performance to optimum will be guaranteed when infinite data are given. Importantly, our proposed BRF is consistent for both classification and regression. The best empirical performance is achieved by BRF when it is compared with state-of-the-art theoretical/consistent RFs. This advance in RF research toward closing the gap between theory and practice is verified by the theoretical and experimental studies in this paper. [ABSTRACT FROM AUTHOR]

In kernel-based classification models, given limited computational power and storage capacity, operations over the full kernel matrix becomes prohibitive. In this paper, we propose a new supervised learning framework using kernel models for sequential data processing. The framework is based on two components that both aim at enhancing the classification capability with a subset selection scheme. The first part is a subspace projection technique in the reproducing kernel Hilbert space using a CLAss-specific Subspace Kernel representation for kernel approximation. In the second part, we propose a novel structural risk minimization algorithm called the adaptive margin slack minimization to iteratively improve the classification accuracy by an adaptive data selection. We motivate each part separately, and then integrate them into learning frameworks for large scale data. We propose two such frameworks: the memory efficient sequential processing for sequential data processing and the parallelized sequential processing for distributed computing with sequential data acquisition. We test our methods on several benchmark data sets and compared with the state-of-the-art techniques to verify the validity of the proposed techniques. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel deep metric learning method to tackle the multi-label image classification problem. In order to better learn the correlations among images features, as well as labels, we attempt to explore a latent space, where images and labels are embedded via two unique deep neural networks, respectively. To capture the relationships between image features and labels, we aim to learn a two-way deep distance metric over the embedding space from two different views, i.e., the distance between one image and its labels is not only smaller than those distances between the image and its labels’ nearest neighbors but also smaller than the distances between the labels and other images corresponding to the labels’ nearest neighbors. Moreover, a reconstruction module for recovering correct labels is incorporated into the whole framework as a regularization term, such that the label embedding space is more representative. Our model can be trained in an end-to-end manner. Experimental results on publicly available image data sets corroborate the efficacy of our method compared with the state of the arts. [ABSTRACT FROM AUTHOR]

Existing synaptic plasticity rules for optimizing the connections between neurons within the reservoir of echo state networks (ESNs) remain to be global in that the same type of plasticity rule with the same parameters is applied to all neurons. However, this is biologically implausible and practically inflexible for learning the structures in the input signals, thereby limiting the learning performance of ESNs. In this paper, we propose to use local plasticity rules that allow different neurons to use different types of plasticity rules and different parameters, which are achieved by optimizing the parameters of the local plasticity rules using the evolution strategy (ES) with covariance matrix adaptation (CMA-ES). We show that evolving neural plasticity will result in a synergistic learning of different plasticity rules, which plays an important role in improving the learning performance. Meanwhile, we show that the local plasticity rules can effectively alleviate synaptic interferences in learning the structure in sensory inputs. The proposed local plasticity rules are compared with a number of the state-of-the-art ESN models and the canonical ESN using a global plasticity rule on a set of widely used prediction and classification benchmark problems to demonstrate its competitive learning performance. [ABSTRACT FROM AUTHOR]

Semisupervised text classification has attracted much attention from the research community. In this paper, a novel model, the semisupervised sequential variational autoencoder (SSVAE), is proposed to tackle this problem. By treating the categorical label of unlabeled data as a discrete latent variable, the proposed model maximizes the variational evidence lower bound of the data likelihood, which implicitly derives the underlying label distribution for the unlabeled data. Analytical work indicates that the autoregressive nature of the sequential model is the crucial issue that renders the vanilla model ineffective. To remedy this, two types of decoders are investigated in the SSVAE model and verified. In addition, a reweighting approach is proposed to circumvent the credit assignment problem that occurs during the reconstruction procedure, which can further improve performance for sparse text data. Experimental results show that our method significantly improves the classification accuracy compared with other modern methods. [ABSTRACT FROM AUTHOR]

This paper presents a nearest neighbor partitioning method designed to improve the performance of a neural-network classifier. For neural-network classifiers, usually the number, positions, and labels of centroids are fixed in partition space before training. However, that approach limits the search for potential neural networks during optimization; the quality of a neural network classifier is based on how clear the decision boundaries are between classes. Although attempts have been made to generate floating centroids automatically, these methods still tend to generate sphere-like partitions and cannot produce flexible decision boundaries. We propose the use of nearest neighbor classification in conjunction with a neural-network classifier. Instead of being bound by sphere-like boundaries (such as the case with centroid-based methods), the flexibility of nearest neighbors increases the chance of finding potential neural networks that have arbitrarily shaped boundaries in partition space. Experimental results demonstrate that the proposed method exhibits superior performance on accuracy and average f-measure. [ABSTRACT FROM AUTHOR]

Feature selection (FS) is an important component of many pattern recognition tasks. In these tasks, one is often confronted with very high-dimensional data. FS algorithms are designed to identify the relevant feature subset from the original features, which can facilitate subsequent analysis, such as clustering and classification. Structured sparsity-inducing feature selection (SSFS) methods have been widely studied in the last few years, and a number of algorithms have been proposed. However, there is no comprehensive study concerning the connections between different SSFS methods, and how they have evolved. In this paper, we attempt to provide a survey on various SSFS methods, including their motivations and mathematical representations. We then explore the relationship among different formulations and propose a taxonomy to elucidate their evolution. We group the existing SSFS methods into two categories, i.e., vector-based feature selection (feature selection based on lasso) and matrix-based feature selection (feature selection based on lr,p -norm). Furthermore, FS has been combined with other machine learning algorithms for specific applications, such as multitask learning, multilabel learning, multiview learning, classification, and clustering. This paper not only compares the differences and commonalities of these methods based on regression and regularization strategies, but also provides useful guidelines to practitioners working in related fields to guide them how to do feature selection. [ABSTRACT FROM AUTHOR]

Convolutional autoencoders (CAEs) have shown their remarkable performance in stacking to deep convolutional neural networks (CNNs) for classifying image data during the past several years. However, they are unable to construct the state-of-the-art CNNs due to their intrinsic architectures. In this regard, we propose a flexible CAE (FCAE) by eliminating the constraints on the numbers of convolutional layers and pooling layers from the traditional CAE. We also design an architecture discovery method by exploiting particle swarm optimization, which is capable of automatically searching for the optimal architectures of the proposed FCAE with much less computational resource and without any manual intervention. We test the proposed approach on four extensively used image classification data sets. Experimental results show that our proposed approach in this paper significantly outperforms the peer competitors including the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

The extreme learning machine (ELM) has attracted much attention over the past decade due to its fast learning speed and convincing generalization performance. However, there still remains a practical issue to be approached when applying the ELM: the randomly generated hidden node parameters without tuning can lead to the hidden node outputs being nonuniformly distributed, thus giving rise to poor generalization performance. To address this deficiency, a novel activation function with an affine transformation (AT) on its input is introduced into the ELM, which leads to an improved ELM algorithm that is referred to as an AT-ELM in this paper. The scaling and translation parameters of the AT activation function are computed based on the maximum entropy principle in such a way that the hidden layer outputs approximately obey a uniform distribution. Application of the AT-ELM algorithm in nonlinear function regression shows its robustness to the range scaling of the network inputs. Experiments on nonlinear function regression, real-world data set classification, and benchmark image recognition demonstrate better performance for the AT-ELM compared with the original ELM, the regularized ELM, and the kernel ELM. Recognition results on benchmark image data sets also reveal that the AT-ELM outperforms several other state-of-the-art algorithms in general. [ABSTRACT FROM AUTHOR]

In this paper, we propose a unified model called flexible affinity matrix learning (FAML) for unsupervised and semisupervised classification by exploiting both the relationship among data and the clustering structure simultaneously. To capture the relationship among data, we exploit the self-expressiveness property of data to learn a structured matrix in which the structures are induced by different norms. A rank constraint is imposed on the Laplacian matrix of the desired affinity matrix, so that the connected components of data are exactly equal to the cluster number. Thus, the clustering structure is explicit in the learned affinity matrix. By making the estimated affinity matrix approximate the structured matrix during the learning procedure, FAML allows the affinity matrix itself to be adaptively adjusted such that the learned affinity matrix can well capture both the relationship among data and the clustering structure. Thus, FAML has the potential to perform better than other related methods. We derive optimization algorithms to solve the corresponding problems. Extensive unsupervised and semisupervised classification experiments on both synthetic data and real-world benchmark data sets show that the proposed FAML consistently outperforms the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Linear discriminant analysis (LDA) is a well-known technique for linear classification, feature extraction, and dimension reduction. To improve the accuracy of LDA under the high dimension low sample size (HDLSS) settings, shrunken estimators, such as Graphical Lasso, can be used to strike a balance between biases and variances. Although the estimator with induced sparsity obtains a faster convergence rate, however, the introduced bias may also degrade the performance. In this paper, we theoretically analyze how the sparsity and the convergence rate of the precision matrix (also known as inverse covariance matrix) estimator would affect the classification accuracy by proposing an analytic model on the upper bound of an LDA misclassification rate. Guided by the model, we propose a novel classifier, $\mathcal {DBSDA}$ , which improves classification accuracy through debiasing. Theoretical analysis shows that $\mathcal {DBSDA}$ possesses a reduced upper bound of misclassification rate and better asymptotic properties than sparse LDA (SDA). We conduct experiments on both synthetic datasets and real application datasets to confirm the correctness of our theoretical analysis and demonstrate the superiority of $\mathcal {DBSDA}$ over LDA, SDA, and other downstream competitors under HDLSS settings. [ABSTRACT FROM AUTHOR]

Partial least squares regression (PLSR) has been a popular technique to explore the linear relationship between two data sets. However, all existing approaches often optimize a PLSR model in Euclidean space and take a successive strategy to calculate all the factors one by one for keeping the mutually orthogonal PLSR factors. Thus, a suboptimal solution is often generated. To overcome the shortcoming, this paper takes statistically inspired modification of PLSR (SIMPLSR) as a representative of PLSR, proposes a novel approach to transform SIMPLSR into optimization problems on Riemannian manifolds, and develops corresponding optimization algorithms. These algorithms can calculate all the PLSR factors simultaneously to avoid any suboptimal solutions. Moreover, we propose sparse SIMPLSR on Riemannian manifolds, which is simple and intuitive. A number of experiments on classification problems have demonstrated that the proposed models and algorithms can get lower classification error rates compared with other linear regression methods in Euclidean space. We have made the experimental code public at https://github.com/Haoran2014. [ABSTRACT FROM AUTHOR]

Many real-world classification problems are characterized by samples of a complex distribution in the input space. The classification accuracy is determined by intrinsic properties of all samples in subspaces of features. This paper proposes a novel algorithm for the construction of radial basis function neural network (RBFNN) classifier based on subspace learning. In this paper, feature subspaces are obtained for every hidden node of the RBFNN during the learning process. The connection weights between the input layer and the hidden layer are adjusted to produce various subspaces with dominative features for different hidden nodes. The network structure and dominative features are encoded in two subpopulations that are cooperatively coevolved using the coevolutionary algorithm to achieve a better global optimality for the estimated RBFNN. Experimental results illustrate that the proposed algorithm is able to obtain RBFNN models with both better classification accuracy and simpler network structure when compared with other learning algorithms. Thus, the proposed model provides a more flexible and efficient approach to complex classification tasks by employing the local characteristics of samples in subspaces. [ABSTRACT FROM PUBLISHER]

Computerized microscopy image analysis plays an important role in computer aided diagnosis and prognosis. Machine learning techniques have powered many aspects of medical investigation and clinical practice. Recently, deep learning is emerging as a leading machine learning tool in computer vision and has attracted considerable attention in biomedical image analysis. In this paper, we provide a snapshot of this fast-growing field, specifically for microscopy image analysis. We briefly introduce the popular deep neural networks and summarize current deep learning achievements in various tasks, such as detection, segmentation, and classification in microscopy image analysis. In particular, we explain the architectures and the principles of convolutional neural networks, fully convolutional networks, recurrent neural networks, stacked autoencoders, and deep belief networks, and interpret their formulations or modelings for specific tasks on various microscopy images. In addition, we discuss the open challenges and the potential trends of future research in microscopy image analysis using deep learning. [ABSTRACT FROM AUTHOR]

Embedding methods have shown promising performance in multilabel prediction, as they are able to discover the label dependence. However, most methods ignore the correlations between the input and output, such that their learned embeddings are not well aligned, which leads to degradation in prediction performance. This paper presents a formulation for multilabel learning, from the perspective of cross-view learning, that explores the correlations between the input and the output. The proposed method, called Co-Embedding (CoE), jointly learns a semantic common subspace and view-specific mappings within one framework. The semantic similarity structure among the embeddings is further preserved, ensuring that close embeddings share similar labels. Additionally, CoE conducts multilabel prediction through the cross-view $k$ nearest neighborhood ($k$ NN) search among the learned embeddings, which significantly reduces computational costs compared with conventional decoding schemes. A hashing-based model, i.e., Co-Hashing (CoH), is further proposed. CoH is based on CoE, and imposes the binary constraint on continuous latent embeddings. CoH aims to generate compact binary representations to improve the prediction efficiency by benefiting from the efficient $k$ NN search of multiple labels in the Hamming space. Extensive experiments on various real-world data sets demonstrate the superiority of the proposed methods over the state of the arts in terms of both prediction accuracy and efficiency. [ABSTRACT FROM AUTHOR]

How to solve the classification problem with only label proportions has recently drawn increasing attention in the machine learning field. In this paper, we propose an ensemble learning strategy to deal with the learning problem with label proportions (LLP). In detail, we first give a loss function based on different weights for LLP, and then construct the corresponding weak classifier, at the same time, estimate its conditional probabilities by a standard logistic function. At last, by introducing the maximum likelihood estimation, we propose a new anyboost learning system for LLP (called Adaboost-LLP). Unlike traditional methods, our method does not make any restrictive assumptions on training set; at the same time, compared with alter- $\propto $ SVM, Adaboost-LLP exploits more extra weight information and uses multiple weak classifiers that can be solved efficiently to combine a strong classifier. All experiments show that our method outperforms the existing methods in both accuracy and training time. [ABSTRACT FROM AUTHOR]

In this paper, we address the semisupervised distance metric learning problem and its applications in classification and image retrieval. First, we formulate a semisupervised distance metric learning model by considering the metric information of inner classes and interclasses. In this model, an adaptive parameter is designed to balance the inner metrics and intermetrics by using data structure. Second, we convert the model to a minimization problem whose variable is symmetric positive-definite matrix. Third, in implementation, we deduce an intrinsic steepest descent method, which assures that the metric matrix is strictly symmetric positive-definite at each iteration, with the manifold structure of the symmetric positive-definite matrix manifold. Finally, we test the proposed algorithm on conventional data sets, and compare it with other four representative methods. The numerical results validate that the proposed method significantly improves the classification with the same computational efficiency. [ABSTRACT FROM AUTHOR]

As data sets become larger and more complicated, an extreme learning machine (ELM) that runs in a traditional serial environment cannot realize its ability to be fast and effective. Although a parallel ELM (PELM) based on MapReduce to process large-scale data shows more efficient learning speed than identical ELM algorithms in a serial environment, some operations, such as intermediate results stored on disks and multiple copies for each task, are indispensable, and these operations create a large amount of extra overhead and degrade the learning speed and efficiency of the PELMs. In this paper, an efficient ELM based on the Spark framework (SELM), which includes three parallel subalgorithms, is proposed for big data classification. By partitioning the corresponding data sets reasonably, the hidden layer output matrix calculation algorithm, matrix \mathbf \hat U decomposition algorithm, and matrix \mathbf V decomposition algorithm perform most of the computations locally. At the same time, they retain the intermediate results in distributed memory and cache the diagonal matrix as broadcast variables instead of several copies for each task to reduce a large amount of the costs, and these actions strengthen the learning ability of the SELM. Finally, we implement our SELM algorithm to classify large data sets. Extensive experiments have been conducted to validate the effectiveness of the proposed algorithms. As shown, our SELM achieves an $8.71\times$ speedup on a cluster with ten nodes, and reaches a $13.79\times$ speedup with 15 nodes, an $18.74\times$ speedup with 20 nodes, a $23.79\times$ speedup with 25 nodes, a $28.89\times$ speedup with 30 nodes, and a $33.81\times$ speedup with 35 nodes. [ABSTRACT FROM AUTHOR]

This paper proposes a novel method, called robust latent subspace learning (RLSL), for image classification. We formulate an RLSL problem as a joint optimization problem over both the latent SL and classification model parameter predication, which simultaneously minimizes: 1) the regression loss between the learned data representation and objective outputs and 2) the reconstruction error between the learned data representation and original inputs. The latent subspace can be used as a bridge that is expected to seamlessly connect the origin visual features and their class labels and hence improve the overall prediction performance. RLSL combines feature learning with classification so that the learned data representation in the latent subspace is more discriminative for classification. To learn a robust latent subspace, we use a sparse item to compensate error, which helps suppress the interference of noise via weakening its response during regression. An efficient optimization algorithm is designed to solve the proposed optimization problem. To validate the effectiveness of the proposed RLSL method, we conduct experiments on diverse databases and encouraging recognition results are achieved compared with many state-of-the-arts methods. [ABSTRACT FROM AUTHOR]

Conventional extreme learning machines (ELMs) solve a Moore–Penrose generalized inverse of hidden layer activated matrix and analytically determine the output weights to achieve generalized performance, by assuming the same loss from different types of misclassification. The assumption may not hold in cost-sensitive recognition tasks, such as face recognition-based access control system, where misclassifying a stranger as a family member may result in more serious disaster than misclassifying a family member as a stranger. Though recent cost-sensitive learning can reduce the total loss with a given cost matrix that quantifies how severe one type of mistake against another, in many realistic cases, the cost matrix is unknown to users. Motivated by these concerns, this paper proposes an evolutionary cost-sensitive ELM, with the following merits: 1) to the best of our knowledge, it is the first proposal of ELM in evolutionary cost-sensitive classification scenario; 2) it well addresses the open issue of how to define the cost matrix in cost-sensitive learning tasks; and 3) an evolutionary backtracking search algorithm is induced for adaptive cost matrix optimization. Experiments in a variety of cost-sensitive tasks well demonstrate the effectiveness of the proposed approaches, with about 5%–10% improvements. [ABSTRACT FROM PUBLISHER]

Batch-mode least squares support vector machine (LSSVM) is often associated with unbounded number of support vectors (SVs’), making it unsuitable for applications involving large-scale streaming data. Limited-scale LSSVM, which allows efficient updating, seems to be a good solution to tackle this issue. In this paper, to train the limited-scale LSSVM dynamically, we present a budget online LSSVM (BOLSSVM) algorithm. Methodologically, by setting a fixed budget for SVs’, we are able to update the LSSVM model according to the updated SVs’ set dynamically without retraining from scratch. In particular, when a new small chunk of SVs’ substitute for the old ones, the proposed algorithm employs a low rank correction technology and the Sherman–Morrison–Woodbury formula to compute the inverse of saddle point matrix derived from the LSSVM’s Karush-Kuhn-Tucker (KKT) system, which, in turn, updates the LSSVM model efficiently. In this way, the proposed BOLSSVM algorithm is especially useful for online prediction tasks. Another merit of the proposed BOLSSVM is that it can be used for k -fold cross validation. Specifically, compared with batch-mode learning methods, the computational complexity of the proposed BOLSSVM method is significantly reduced from \mathcal {O}(n^{4}) to \mathcal {O}(n^{3})$ for leave-one-out cross validation with $n$ training samples. The experimental results of classification and regression on benchmark data sets and real-world applications show the validity and effectiveness of the proposed BOLSSVM algorithm. [ABSTRACT FROM AUTHOR]

This paper tries to separate fine-grained images by jointly learning the encoding parameters and codebooks through low-rank sparse coding (LRSC) with general and class-specific codebook generation. Instead of treating each local feature independently, we encode the local features within a spatial region jointly by LRSC. This ensures that the spatially nearby local features with similar visual characters are encoded by correlated parameters. In this way, we can make the encoded parameters more consistent for fine-grained image representation. Besides, we also learn a general codebook and a number of class-specific codebooks in combination with the encoding scheme. Since images of fine-grained classes are visually similar, the difference is relatively small between the general codebook and each class-specific codebook. We impose sparsity constraints to model this relationship. Moreover, the incoherences with different codebooks and class-specific codebooks are jointly considered. We evaluate the proposed method on several public image data sets. The experimental results show that by learning general and class-specific codebooks with the joint encoding of local features, we are able to model the differences among different fine-grained classes than many other fine-grained image classification methods. [ABSTRACT FROM AUTHOR]

Residual classifiers are common in dictionary-based multiclass classification. This paper proposes the concept of performance functions for residual classifiers. A performance function for multiclass classifications is a conceptual measurement function that combines local and global measurements. In general, the performance function is nonlinear. To explore the properties of the performance function, we employ the Taylor series expansion technique and derive a family of measurement functions. Specifically, the linear measurement and the quadratic measurement (QM) are derived. By exploiting the effect of the higher order terms in the performance function as well as the fundamental nondecreasing constrain, we derive the normalized QM (NQM). We present the classifier for multiclass classification using the proposed measurements. The proposed algorithms are tested against frontal faces and handwritten digit recognition tasks. Our tests show that the QM classifier achieves competitive classification results compared with baseline methods. NQM shows better stability with different parameter configurations. [ABSTRACT FROM AUTHOR]

The sparse representation-based classification (SRC) has been utilized in many applications and is an effective algorithm in machine learning. However, the performance of SRC highly depends on the data distribution. Some existing works proved that SRC could not obtain satisfactory results on uncontrolled data sets. Except the uncontrolled data sets, SRC cannot deal with imbalanced classification either. In this paper, we proposed a model named sparse supervised representation classifier (SSRC) to solve the above-mentioned issues. The SSRC involves the class label information during the test sample representation phase to deal with the uncontrolled data sets. In SSRC, each class has the opportunity to linearly represent the test sample in its subspace, which can decrease the influences of the uncontrolled data distribution. In order to classify imbalanced data sets, a class weight learning model is proposed and added to SSRC. Each class weight is learned from its corresponding training samples. The experimental results based on the AR face database (uncontrolled) and 15 KEEL data sets (imbalanced) with an imbalanced rate ranging from 1.48 to 61.18 prove SSRC can effectively classify uncontrolled and imbalanced data sets. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel strategy to combine a set of compact descriptors to leverage an associated recognition task. We formulate the problem from a multiple kernel learning (MKL) perspective and solve it following a stochastic variance reduced gradient (SVRG) approach to address its scalability, currently an open issue. MKL models are ideal candidates to jointly learn the optimal combination of features along with its associated predictor. However, they are unable to scale beyond a dozen thousand of samples due to high computational and memory requirements, which severely limits their applicability. We propose SVRG-MKL, an MKL solution with inherent scalability properties that can optimally combine multiple descriptors involving millions of samples. Our solution takes place directly in the primal to avoid Gram matrices computation and memory allocation, whereas the optimization is performed with a proposed algorithm of linear complexity and hence computationally efficient. Our proposition builds upon recent progress in SVRG with the distinction that each kernel is treated differently during optimization, which results in a faster convergence than applying off-the-shelf SVRG into MKL. Extensive experimental validation conducted on several benchmarking data sets confirms a higher accuracy and a significant speedup of our solution. Our technique can be extended to other MKL problems, including visual search and transfer learning, as well as other formulations, such as group-sensitive (GMKL) and localized MKL (LMKL) in convex settings. [ABSTRACT FROM AUTHOR]

The $k$ -nearest neighbor (KNN) rule is a successful technique in pattern classification due to its simplicity and effectiveness. As a supervised classifier, KNN classification performance usually suffers from low-quality samples in the training data set. Thus, training data set cleaning (TDC) methods are needed for enhancing the classification accuracy by cleaning out noisy, or even wrong, samples in the original training data set. In this paper, we propose a classification ability ranking (CAR)-based TDC method to improve the performance of a KNN classifier, namely CAR-based TDC method. The proposed classification ability function ranks a training sample in terms of its contribution to correctly classify other training samples as a KNN through the leave-one-out (LV1) strategy in the cleaning stage. The training sample that likely misclassifies the other samples during the KNN classifications according to the LV1 strategy is considered to have lower classification ability and will be cleaned out from the original training data set. Extensive experiments, based on ten real-world data sets, show that the proposed CAR-based TDC method can significantly reduce the classification error rates of KNN-based classifiers, while reducing computational complexity thanks to a smaller cleaned training data set. [ABSTRACT FROM AUTHOR]

Linear dynamical system (LDS) offers a convenient way to reveal the unobservable structure behind the data. This makes it useful for data representation and explanatory analysis. An immediate limitation with this model is that most training algorithms train a model to best approximate a sequential instance. They do not consider its class or label which indicates the dissimilarity/similarity to other instances. As a result, LDS’s trained in this way are inclined to be indistinguishable over classes, resulting in a poor performance in the model-based classification. In this paper, after revisiting this limitation, we propose to promote the diversity between the two models of different classes. The diversity, measured by determinantal point process (DPP) on LDS’s, is utilized to remedy the greedy behavior of the electromagnetic algorithm. The training goal is a model that balances the goodness of fit and being distinguishable over classes. Experiments on synthetic data confirm its effectiveness in generating discriminative systems under supervisory information. The classification on short time-span data sets confirms that the models generated by our approach could generalize well to unseen data. [ABSTRACT FROM AUTHOR]

This paper presents several L1 -minimization algorithms for sparse signal reconstruction based on a continuous-time projection neural network (PNN). First, a one-layer projection neural network is designed based on a projection operator and a projection matrix. The stability and global convergence of the proposed neural network are proved. Then, based on a discrete-time version of the PNN, several L1 -minimization algorithms for sparse signal reconstruction are developed and analyzed. Experimental results based on random Gaussian sparse signals show the effectiveness and performance of the proposed algorithms. Moreover, experimental results based on two face image databases are presented that reveal the influence of sparsity to the recognition rate. The algorithms are shown to be robust to the amplitude and sparsity level of signals as well as efficient with high convergence rate compared with several existing L1 -minimization algorithms. [ABSTRACT FROM PUBLISHER]

The one-class classification problem is a well-known research endeavor in pattern recognition. The problem is also known under different names, such as outlier and novelty/anomaly detection. The core of the problem consists in modeling and recognizing patterns belonging only to a so-called target class. All other patterns are termed nontarget, and therefore, they should be recognized as such. In this paper, we propose a novel one-class classification system that is based on an interplay of different techniques. Primarily, we follow a dissimilarity representation-based approach; we embed the input data into the dissimilarity space (DS) by means of an appropriate parametric dissimilarity measure. This step allows us to process virtually any type of data. The dissimilarity vectors are then represented by weighted Euclidean graphs, which we use to determine the entropy of the data distribution in the DS and at the same time to derive effective decision regions that are modeled as clusters of vertices. Since the dissimilarity measure for the input data is parametric, we optimize its parameters by means of a global optimization scheme, which considers both mesoscopic and structural characteristics of the data represented through the graphs. The proposed one-class classifier is designed to provide both hard (Boolean) and soft decisions about the recognition of test patterns, allowing an accurate description of the classification process. We evaluate the performance of the system on different benchmarking data sets, containing either feature-based or structured patterns. Experimental results demonstrate the effectiveness of the proposed technique. [ABSTRACT FROM PUBLISHER]

Based on a generative adversarial network (GAN), a novel version named Task-Oriented GAN is proposed to tackle difficulties in PolSAR image interpretation, including PolSAR data analysis and small sample problem. Besides two typical parts in GAN, i.e., generator (G-Net) and discriminator (D-Net), there is a third part named TaskNet (T-Net) in the Task-Oriented GAN, where T-Net is employed to accomplish a certain task. Two tasks, PolSAR image classification and clustering, are studied in this paper, where T-Net acts as a Classifier and a Clusterer, respectively. The learning procedure of Task-Oriented GAN consists of two main stages. In the first stage, G-Net and D-Net vie with each other like that in a general GAN; in the second stage, G-Net is adjusted and oriented by T-Net so that more samples, which are benefit for the task and called fake data, are generated. As a result, Task-Oriented GAN not only has the advantage of GAN (no-assumption data modeling) but also overcomes the disadvantage of GAN (task-free). After learning, fake data are employed to enrich training set and avoid overfitting; so Task-Oriented GAN performs well even if the manual-labeled data are small. To verify the effectiveness of T-Net, a visualized comparison is provided, where some fake digits generated from Task-Oriented GAN are illustrated along with that from GAN. What is more, considering that there is a great difference between PolSAR data and general data, in our PolSAR image classification and clustering tasks, the specific PolSAR information is inserted into the structure of the Task-Oriented GAN. This enables researchers to mine inherent information in PolSAR data without any data hypothesis and find ways for small sample problem at the same time. Experiment results tested on three PolSAR images show that the proposed method performs well in dealing with PolSAR image classification and clustering. [ABSTRACT FROM AUTHOR]

Boosting is a learning scheme that combines weak learners to produce a strong composite learner, with the underlying intuition that one can obtain accurate learner by combining “rough” ones. This paper aims at developing a new boosting strategy, called rescaled boosting (RBoosting), to accelerate the numerical convergence rate and, consequently, improve learning performances of the original boosting. Our studies show that RBoosting possesses the almost optimal numerical convergence rate in the sense that, up to a logarithmic factor, it can reach the minimax nonlinear approximation rate. We then use RBoosting to tackle classification problems and deduce corresponding statistical consistency and tight generalization error estimates. A series of theoretical and experimental results shows that RBoosting outperforms boosting in terms of generalization. [ABSTRACT FROM AUTHOR]

The paper presents a new framework for complex support vector regression (SVR) as well as Support Vector Machines (SVM) for quaternary classification. The method exploits the notion of widely linear estimation to model the input-out relation for complex-valued data and considers two cases: 1) the complex data are split into their real and imaginary parts and a typical real kernel is employed to map the complex data to a complexified feature space and 2) a pure complex kernel is used to directly map the data to the induced complex feature space. The recently developed Wirtinger’s calculus on complex reproducing kernel Hilbert spaces is employed to compute the Lagrangian and derive the dual optimization problem. As one of our major results, we prove that any complex SVM/SVR task is equivalent with solving two real SVM/SVR tasks exploiting a specific real kernel, which is generated by the chosen complex kernel. In particular, the case of pure complex kernels leads to the generation of new kernels, which have not been considered before. In the classification case, the proposed framework inherently splits the complex space into four parts. This leads naturally to solving the four class-task (quaternary classification), instead of the typical two classes of the real SVM. In turn, this rationale can be used in a multiclass problem as a split-class scenario based on four classes, as opposed to the one-versus-all method; this can lead to significant computational savings. Experiments demonstrate the effectiveness of the proposed framework for regression and classification tasks that involve complex data. [ABSTRACT FROM AUTHOR]

In this paper, a new method for constructing decision trees for stream data is proposed. First a new splitting criterion based on the misclassification error is derived. A theorem is proven showing that the best attribute computed in considered node according to the available data sample is the same, with some high probability, as the attribute derived from the whole infinite data stream. Next this result is combined with the splitting criterion based on the Gini index. It is shown that such combination provides the highest accuracy among all studied algorithms. [ABSTRACT FROM PUBLISHER]

Well known for its simplicity and effectiveness in classification, AdaBoost, however, suffers from overfitting when class-conditional distributions have significant overlap. Moreover, it is very sensitive to noise that appears in the labels. This paper tackles the above limitations simultaneously via optimizing a modified loss function (i.e., the conditional risk). The proposed approach has the following two advantages. First, it is able to directly take into account label uncertainty with an associated label confidence. Second, it introduces a trustworthiness measure on training samples via the Bayesian risk rule, and hence the resulting classifier tends to have finite sample performance that is superior to that of the original AdaBoost when there is a large overlap between class conditional distributions. Theoretical properties of the proposed method are investigated. Extensive experimental results using synthetic data and real-world data sets from UCI machine learning repository are provided. The empirical study shows the high competitiveness of the proposed method in predication accuracy and robustness when compared with the original AdaBoost and several existing robust AdaBoost algorithms. [ABSTRACT FROM AUTHOR]

This paper presents an enhanced rank-order-based learning algorithm, called SpikeTemp, for spiking neural networks (SNNs) with a dynamically adaptive structure. The trained feed-forward SNN consists of two layers of spiking neurons: 1) an encoding layer which temporally encodes real-valued features into spatio-temporal spike patterns and 2) an output layer of dynamically grown neurons which perform spatio-temporal classification. Both Gaussian receptive fields and square cosine population encoding schemes are employed to encode real-valued features into spatio-temporal spike patterns. Unlike the rank-order-based learning approach, SpikeTemp uses the precise times of the incoming spikes for adjusting the synaptic weights such that early spikes result in a large weight change and late spikes lead to a smaller weight change. This removes the need to rank all the incoming spikes and, thus, reduces the computational cost of SpikeTemp. The proposed SpikeTemp algorithm is demonstrated on several benchmark data sets and on an image recognition task. The results show that SpikeTemp can achieve better classification performance and is much faster than the existing rank-order-based learning approach. In addition, the number of output neurons is much smaller when the square cosine encoding scheme is employed. Furthermore, SpikeTemp is benchmarked against a selection of existing machine learning algorithms, and the results demonstrate the ability of SpikeTemp to classify different data sets after just one presentation of the training samples with comparable classification performance. [ABSTRACT FROM PUBLISHER]

It has always been a challenging task to develop a fast and an efficient incremental linear discriminant analysis (ILDA) algorithm. For this purpose, we conduct a new study for linear discriminant analysis (LDA) in this paper and develop a new ILDA algorithm. We propose a new batch LDA algorithm called LDA/QR. LDA/QR is a simple and fast LDA algorithm, which is obtained by computing the economic QR factorization of the data matrix followed by solving a lower triangular linear system. The relationship between LDA/QR and uncorrelated LDA (ULDA) is also revealed. Based on LDA/QR, we develop a new incremental LDA algorithm called ILDA/QR. The main features of our ILDA/QR include that: 1) it can easily handle the update from one new sample or a chunk of new samples; 2) it has efficient computational complexity and space complexity; and 3) it is very fast and always achieves competitive classification accuracy compared with ULDA algorithm and existing ILDA algorithms. Numerical experiments based on some real-world data sets demonstrate that our ILDA/QR is very efficient and competitive with the state-of-the-art ILDA algorithms in terms of classification accuracy, computational complexity, and space complexity. [ABSTRACT FROM PUBLISHER]

We investigate classification and regression for nonuniformly sampled variable length sequential data and introduce a novel long short-term memory (LSTM) architecture. In particular, we extend the classical LSTM network with additional time gates, which incorporate the time information as a nonlinear scaling factor on the conventional gates. We also provide forward-pass and backward-pass update equations for the proposed LSTM architecture. We show that our approach is superior to the classical LSTM architecture when there is correlation between time samples. In our experiments, we achieve significant performance gains with respect to the classical LSTM and phased-LSTM architectures. In this sense, the proposed LSTM architecture is highly appealing for the applications involving nonuniformly sampled sequential data. [ABSTRACT FROM AUTHOR]

Support vector machines (SVMs) seek to optimize three distinct objectives: maximization of margin, minimization of regularization from the positive class, and minimization of regularization from the negative class. The right choice of weightage for each of these objectives is critical to the quality of the classifier learned, especially in case of the class imbalanced data sets. Therefore, costly parameter tuning has to be undertaken to find a set of suitable relative weights. In this brief, we propose to train SVMs, on two-class as well as multiclass data sets, in a multiobjective optimization framework called radial boundary intersection to overcome this shortcoming. The experimental results suggest that the radial boundary intersection-based scheme is indeed effective in finding the best tradeoff among the objectives compared with parameter-tuning schemes. [ABSTRACT FROM AUTHOR]

An artificial neural network (ANN) that mimics the information processing mechanisms and procedures of neurons in human brains has achieved a great success in many fields, e.g., classification, prediction, and control. However, traditional ANNs suffer from many problems, such as the hard understanding problem, the slow and difficult training problems, and the difficulty to scale them up. These problems motivate us to develop a new dendritic neuron model (DNM) by considering the nonlinearity of synapses, not only for a better understanding of a biological neuronal system, but also for providing a more useful method for solving practical problems. To achieve its better performance for solving problems, six learning algorithms including biogeography-based optimization, particle swarm optimization, genetic algorithm, ant colony optimization, evolutionary strategy, and population-based incremental learning are for the first time used to train it. The best combination of its user-defined parameters has been systemically investigated by using the Taguchi’s experimental design method. The experiments on 14 different problems involving classification, approximation, and prediction are conducted by using a multilayer perceptron and the proposed DNM. The results suggest that the proposed learning algorithms are effective and promising for training DNM and thus make DNM more powerful in solving classification, approximation, and prediction problems. [ABSTRACT FROM AUTHOR]

The problem of early classification of time series appears naturally in contexts where the data, of temporal nature, are collected over time, and early class predictions are interesting or even required. The objective is to classify the incoming sequence as soon as possible, while maintaining suitable levels of accuracy in the predictions. Thus, we can say that the problem of early classification consists of optimizing two objectives simultaneously: accuracy and earliness. In this context, we present a method for early classification based on combining a set of probabilistic classifiers together with a stopping rule (SR). This SR will act as a trigger and will tell us when to output a prediction or when to wait for more data, and its main novelty lies in the fact that it is built by explicitly optimizing a cost function based on accuracy and earliness. We have selected a large set of benchmark data sets and four other state-of-the-art early classification methods, and we have evaluated and compared our framework obtaining superior results in terms of both earliness and accuracy. [ABSTRACT FROM AUTHOR]

This brief proposes a truncated \ell 1 distance (TL1) kernel, which results in a classifier that is nonlinear in the global region but is linear in each subregion. With this kernel, the subregion structure can be trained using all the training data and local linear classifiers can be established simultaneously. The TL1 kernel has good adaptiveness to nonlinearity and is suitable for problems which require different nonlinearities in different areas. Though the TL1 kernel is not positive semidefinite, some classical kernel learning methods are still applicable which means that the TL1 kernel can be directly used in standard toolboxes by replacing the kernel evaluation. In numerical experiments, the TL1 kernel with a pregiven parameter achieves similar or better performance than the radial basis function kernel with the parameter tuned by cross validation, implying the TL1 kernel a promising nonlinear kernel for classification tasks. [ABSTRACT FROM AUTHOR]

To deal with ambiguities in partial label learning (PLL), the existing PLL methods implement disambiguations, by either identifying the ground-truth label or averaging the candidate labels. However, these methods can be easily misled by the false-positive labels in the candidate label set. We find that these ambiguities often originate from the noise caused by highly correlated or overlapping candidate labels, which leads to the difficulty in identifying the ground-truth label on the first attempt. To give the trained models more tolerance, we first propose the top- $k$ partial loss and convex top- $k$ partial hinge loss. Based on the losses, we present a novel top- $k$ partial label machine (TPLM) for partial label classification. An efficient optimization algorithm is proposed based on accelerated proximal stochastic dual coordinate ascent (Prox-SDCA) and linear programming (LP). Moreover, we present a theoretical analysis of the generalization error for TPLM. Comprehensive experiments on both controlled UCI datasets and real-world partial label datasets demonstrate that the proposed method is superior to the state-of-the-art approaches. [ABSTRACT FROM AUTHOR]