Search

Tensor singular value decomposition (t-SVD) has recently become increasingly popular for tensor recovery under partial and/or corrupted observations. However, the existing t -SVD-based methods neither make use of a rank prior nor provide an accurate rank estimation (RE), which would limit their recovery performance. From the practical perspective, the tensor RE problem is nontrivial and difficult to solve. In this article, we, therefore, aim to determine the correct rank of an intrinsic low-rank tensor from corrupted observations based on t-SVD and further improve recovery results with the estimated rank. Specifically, we first induce the equivalence of the tensor nuclear norm (TNN) of a tensor and its f -diagonal tensor. We then simultaneously minimize the reconstruction error and TNN of the f -diagonal tensor, leading to RE. Subsequently, we relax our model by removing the TNN regularizer to improve the recovery performance. Furthermore, we consider more general cases in the presence of missing data and/or gross corruptions by proposing robust tensor principal component analysis and robust tensor completion with RE. The robust methods can achieve successful recovery by refining the models with correct estimated ranks. Experimental results show that the proposed methods outperform the state-of-the-art methods with significant improvements.

Existing transfer learning methods that focus on problems in stationary environments are not usually applicable to dynamic environments, where concept drift may occur. To the best of our knowledge, the concept drift-tolerant transfer learning (CDTL), whose major challenge is the need to adapt the target model and knowledge of source domains to the changing environments, has yet to be well explored in the literature. This article, therefore, proposes a hybrid ensemble approach to deal with the CDTL problem provided that data in the target domain are generated in a streaming chunk-by-chunk manner from nonstationary environments. At each time step, a class-wise weighted ensemble is presented to adapt the model of target domains to new environments. It assigns a weight vector for each classifier generated from the previous data chunks to allow each class of the current data leveraging historical knowledge independently. Then, a domain-wise weighted ensemble is introduced to combine the source and target models to select useful knowledge of each domain. The source models are updated with the source instances performed by the proposed adaptive weighted CORrelation ALignment (AW-CORAL). AW-CORAL iteratively minimizes domain discrepancy meanwhile decreases the effect of unrelated source instances. In this way, positive knowledge of source domains can be potentially promoted while negative knowledge is reduced. Empirical studies on synthetic and real benchmark data sets demonstrate the effectiveness of the proposed algorithm.

Long-tailed data is still a big challenge for deep neural networks, even though they have achieved great success on balanced data. We observe that vanilla training on long-tailed data with cross-entropy loss makes the instance-rich head classes severely squeeze the spatial distribution of the tail classes, which leads to difficulty in classifying tail class samples. Furthermore, the original cross-entropy loss can only propagate gradient short-lively because the gradient in softmax form rapidly approaches zero as the logit difference increases. This phenomenon is called softmax saturation. It is unfavorable for training on balanced data, but can be utilized to adjust the validity of the samples in long-tailed data, thereby solving the distorted embedding space of long-tailed problems. To this end, this paper proposes the Gaussian clouded logit adjustment by Gaussian perturbation of different class logits with varied amplitude. We define the amplitude of perturbation as cloud size and set relatively large cloud sizes to tail classes. The large cloud size can reduce the softmax saturation and thereby making tail class samples more active as well as enlarging the embedding space. To alleviate the bias in a classifier, we therefore propose the class-based effective number sampling strategy with classifier re-training. Extensive experiments on benchmark datasets validate the superior performance of the proposed method. Source code is available at https://github.com/Keke921/GCLLoss., Published as a conference paper at CVPR 2022. arXiv admin note: text overlap with arXiv:2305.10648

Password guessing is an important issue in user security and privacy protection. Using generative adversarial network (GAN) to guess passwords is a new strategy emerging in recent years, which exploits the discriminator’s evaluation of passwords to guide the update of the generator so that password guessing sets can be produced. However, the sampling process of discrete data from a categorical distribution is not differentiable so that backpropagation does not work well. In this paper, we propose a novel password guessing model named G-Pass, which consists of two main components. The first is a new network structure, which modifies the generator from the convolutional neural network (CNN) to long short-term memory- (LSTM-) based network and employs multiple convolutional layers in the discriminator to provide more informative signals for generator updating. The second is Gumbel-Softmax with temperature control for training GAN on passwords. Experimental results show the proposed G-Pass outperforms PassGAN in password quality and cracking rate. Moreover, by dynamically adjusting one parameter during the training process, a trade-off between sample diversity and quality can be achieved with our proposed model.

Ordinal attribute has all the common characteristics of a nominal one but it differs from the nominal one by having naturally ordered possible values (also called categories interchangeably). In clustering analysis tasks, categorical data composed of both ordinal and nominal attributes (also called mixed-categorical data interchangeably) are common. Under this circumstance, existing distance and similarity measures suffer from at least one of the following two drawbacks: 1) directly treat ordinal attributes as nominal ones, and thus ignore the order information from them and 2) suppose all the attributes are independent of each other, measure the distance between two categories from a target attribute without considering the valuable information provided by the other attributes that correlate with the target one. These two drawbacks may twist the natural distances of attributes and further lead to unsatisfactory clustering results. This article, therefore, presents an entropy-based distance metric that quantifies the distance between categories by exploiting the information provided by different attributes that correlate with the target one. It also preserves the order relationship among ordinal categories during the distance measurement. Since attributes are usually correlated in different degrees, we also define the interdependence between different types of attributes to weight their contributions in forming distances. The proposed metric overcomes the two above-mentioned drawbacks for mixed-categorical data clustering. More important, it conceptually unifies the distances of ordinal and nominal attributes to avoid information loss during clustering. Moreover, it is parameter free, and will not bring extra computational cost compared to the existing state-of-the-art counterparts. Extensive experiments show the superiority of the proposed distance metric.

Fine-grained image-text retrieval has been a hot research topic to bridge the vision and languages, and its main challenge is how to learn the semantic correspondence across different modalities. The existing methods mainly focus on learning the global semantic correspondence or intramodal relation correspondence in separate data representations, but which rarely consider the intermodal relation that interactively provide complementary hints for fine-grained semantic correlation learning. To address this issue, we propose a relation-aggregated cross-graph (RACG) model to explicitly learn the fine-grained semantic correspondence by aggregating both intramodal and intermodal relations, which can be well utilized to guide the feature correspondence learning process. More specifically, we first build semantic-embedded graph to explore both fine-grained objects and their relations of different media types, which aim not only to characterize the object appearance in each modality, but also to capture the intrinsic relation information to differentiate intramodal discrepancies. Then, a cross-graph relation encoder is newly designed to explore the intermodal relation across different modalities, which can mutually boost the cross-modal correlations to learn more precise intermodal dependencies. Besides, the feature reconstruction module and multihead similarity alignment are efficiently leveraged to optimize the node-level semantic correspondence, whereby the relation-aggregated cross-modal embeddings between image and text are discriminatively obtained to benefit various image-text retrieval tasks with high retrieval performance. Extensive experiments evaluated on benchmark datasets quantitatively and qualitatively verify the advantages of the proposed framework for fine-grained image-text retrieval and show its competitive performance with the state of the arts.

This work addresses unsupervised partial domain adaptation (PDA), in which classes in the target domain are a subset of the source domain. The key challenges of PDA are how to leverage source samples in the shared classes to promote positive transfer and filter out the irrelevant source samples to mitigate negative transfer. Existing PDA methods based on adversarial DA do not consider the loss of class discriminative representation. To this end, this article proposes a contrastive learning-assisted alignment (CLA) approach for PDA to jointly align distributions across domains for better adaptation and to reweight source instances to reduce the contribution of outlier instances. A contrastive learning-assisted conditional alignment (CLCA) strategy is presented for distribution alignment. CLCA first exploits contrastive losses to discover the class discriminative information in both domains. It then employs a contrastive loss to match the clusters across the two domains based on adversarial domain learning. In this respect, CLCA attempts to reduce the domain discrepancy by matching the class-conditional and marginal distributions. Moreover, a new reweighting scheme is developed to improve the quality of weights estimation, which explores information from both the source and the target domains. Empirical results on several benchmark datasets demonstrate that the proposed CLA outperforms the existing state-of-the-art PDA methods.

Heterogeneous attribute data composed of attributes with different types of values are quite common in a variety of real-world applications. As data annotation is usually expensive, clustering has provided a promising way for processing unlabeled data, where the adopted similarity measure plays a key role in determining the clustering accuracy. However, it is a very challenging task to appropriately define the similarity between data objects with heterogeneous attributes because the values from heterogeneous attributes are generally with very different characteristics. Specifically, numerical attributes are with quantitative values, while categorical attributes are with qualitative values. Furthermore, categorical attributes can be categorized into nominal and ordinal ones according to the order information of their values. To circumvent the awkward gap among the heterogeneous attributes, this article will propose a new dissimilarity metric for cluster analysis of such data. We first study the connections among the heterogeneous attributes and build graph representations for them. Then, a metric is proposed, which computes the dissimilarities between attribute values under the guidance of the graph structures. Finally, we develop a new k -means-type clustering algorithm associated with this proposed metric. It turns out that the proposed method is competent to perform cluster analysis of datasets composed of an arbitrary combination of numerical, nominal, and ordinal attributes. Experimental results show its efficacy in comparison with its counterparts.

Ship detection plays a significant role in the high-resolution remote sensing (HRRS) community, but it is a challenging task due to the complex contents within HRRS images and the diverse orientation of ships. Recently, with the development of deep learning, the performance of the HRRS ship detection model has been improved greatly. Most of them employ deep networks and complicate anchor mechanism to get well ship detection results. Nevertheless, this kind of combination limits the detection efficiency. To address this problem, a new approach named accurate and real-time rotational ship detector (AR²Det) is proposed in this article to detect ships without the anchor mechanism. Based on the extracted features by the feature extraction module (FEM) and the central information of ships, AR²Det adopts two simple modules, ship detector (SDet) and center detector (CDet), to generate and improve the detection results, respectively. AR²Det is efficient due to the simple postprocessing and the lightweight network. Also, AR²Det performs satisfactorily due to the effective generation and enhancement strategy of bounding boxes. The extensive experiments are conducted on a public HRRS image ship detection dataset HRSC2016. The promising results show that our method outperforms the state-of-the-art approaches in terms of both accuracy and speed.

Despite the great success of the existing work in fine-grained visual categorization (FGVC), there are still several unsolved challenges, e.g., poor interpretation and vagueness contribution. To circumvent this drawback, motivated by the hypersphere embedding method, we propose a discriminative suprasphere embedding (DSE) framework, which can provide intuitive geometric interpretation and effectively extract discriminative features. Specifically, DSE consists of three modules. The first module is a suprasphere embedding (SE) block, which learns discriminative information by emphasizing weight and phase. The second module is a phase activation map (PAM) used to analyze the contribution of local descriptors to the suprasphere feature representation, which uniformly highlights the object region and exhibits remarkable object localization capability. The last module is a class contribution map (CCM), which quantitatively analyzes the network classification decision and provides insight into the domain knowledge about classified objects. Comprehensive experiments on three benchmark datasets demonstrate the effectiveness of our proposed method in comparison with state-of-the-art methods.

For long-tailed distributed data, existing classification models often learn overwhelmingly on the head classes while ignoring the tail classes, resulting in poor generalization capability. To address this problem, we thereby propose a new approach in this paper, in which a key point sensitive (KPS) loss is presented to regularize the key points strongly to improve the generalization performance of the classification model. Meanwhile, in order to improve the performance on tail classes, the proposed KPS loss also assigns relatively large margins on tail classes. Furthermore, we propose a gradient adjustment (GA) optimization strategy to re-balance the gradients of positive and negative samples for each class. By virtue of the gradient analysis of the loss function, it is found that the tail classes always receive negative signals during training, which misleads the tail prediction to be biased towards the head. The proposed GA strategy can circumvent excessive negative signals on tail classes and further improve the overall classification accuracy. Extensive experiments conducted on long-tailed benchmarks show that the proposed method is capable of significantly improving the classification accuracy of the model in tail classes while maintaining competent performance in head classes.

Image-text matching of natural scenes has been a popular research topic in both computer vision and natural language processing communities. Recently, fine-grained image-text matching has shown its significant advance in inferring the high-level semantic correspondence by aggregating pairwise region-word similarity, but it remains challenging mainly due to insufficient representation of high-order semantic concepts and their explicit connections in one modality as its matched in another modality. To tackle this issue, we propose a relationship-enhanced semantic graph (ReSG) model, which can improve the image-text representations by learning their locally discriminative semantic concepts and then organizing their relationships in a contextual order. To be specific, two tailored graph encoders, visual relationship-enhanced graph (VReG) and textual relationship-enhanced graph (TReG), are respectively exploited to encode the high-level semantic concepts of corresponding instances and their semantic relationships. Meanwhile, the representations of each graph node are optimized by aggregating semantically contextual information to enhance the node-level semantic correspondence. Further, the hard-negative triplet ranking loss, center hinge loss, and positive-negative margin loss are jointly leveraged to learn the fine-grained correspondence between the ReSG representations of image and text, whereby the discriminative cross-modal embeddings can be explicitly obtained to benefit various image-text matching tasks in a more interpretable way. Extensive experiments verify the advantages of the proposed fine-grained graph matching approach, by achieving the state-of-the-art image-text matching results on public benchmark datasets.

In recent years, with the development of deep learning (DL), the hyperspectral image (HSI) classification methods based on DL have shown superior performance. Although these DL-based methods have great successes, there is still room to improve their ability to explore spatial–spectral information. In this article, we propose a 3-D octave convolution with the spatial–spectral attention network (3DOC-SSAN) to capture discriminative spatial–spectral features for the classification of HSIs. Especially, we first extend the octave convolution model using 3-D convolution, namely, a 3-D octave convolution model (3D-OCM), in which four 3-D octave convolution blocks are combined to capture spatial–spectral features from HSIs. Not only the spatial information can be mined deeply from the high- and low-frequency aspects but also the spectral information can be taken into account by our 3D-OCM. Second, we introduce two attention models from spatial and spectral dimensions to highlight the important spatial areas and specific spectral bands that consist of significant information for the classification tasks. Finally, in order to integrate spatial and spectral information, we design an information complement model to transmit important information between spatial and spectral attention features. Through the information complement model, the beneficial parts of spatial and spectral attention features for the classification tasks can be fully utilized. Comparing with several existing popular classifiers, our proposed method can achieve competitive performance on four benchmark data sets.

Due to the high dimensionality of hyperspectral images (HSIs), more training samples are needed in general for better classification performance. However, surface materials cannot always provide sufficient training samples in practice. HSI classification with small size training samples is still a challenging problem. Multiview learning is a feasible way to improve the classification accuracy in the case of small training samples by combining information from different views. This article proposes a new spatial window-based multiview intact feature learning method (SWMIFL) for HSI classification. In the proposed SWMIFL, multiple features that reflect different information of the original image are extracted and spatial windows are imposed on training samples to select unlabeled samples. Then, multiview intact feature learning is performed to learn the intact feature of the training and unlabeled samples. Considering that neighboring samples are likely to belong to the same class, labels of spatial neighboring samples are determined by two factors including the labels of training samples that locate in the spatial window and the labels learned from the intact feature. Finally, unlabeled samples that have same labels under these two factors are treated as new training samples. Experimental results demonstrate that the proposed SWMIFL-based classification method outperforms several well-known HSI classification methods on three real-world data sets.

The papers in this special section focus on emerging computational intelligent techniques for decision making with Big Data in uncertain environments. Decision making in a big-data environment poses many challenges because of the high dimensional, heterogeneous, complex, unstructured, and unpredictable characteristics of the data which often suffer from different kinds of uncertainty. The uncertainty in the data may arise due to many factors including missing values, imprecise measurements, changes in process characteristics during the data generation period, lack of appropriate monitoring of data measurement process to name a few. Internet-of-Things (IoT) systems usually generate a large amount of unstructured and heterogeneous data demanding specialized techniques for data analytics. Thus, decision making in such an environment poses significant challenges and often demands new and innovative design techniques and algorithms for decision making.

Reversible data hiding in ciphertext has potential applications for privacy protection and transmitting extra data in a cloud environment. For instance, an original plain-text image can be recovered from the encrypted image generated after data embedding, while the embedded data can be extracted before or after decryption. However, homomorphic processing can hardly be applied to an encrypted image with hidden data to generate the desired image. This is partly due to that the image content may be changed by preprocessing or/and data embedding. Even if the corresponding plain-text pixel values are kept unchanged by lossless data hiding, the hidden data will be destroyed by outer processing. To address this issue, a lossless data hiding method called random element substitution (RES) is proposed for the Paillier cryptosystem by substituting the to-be-hidden bits for the random element of a cipher value. Moreover, the RES method is combined with another preprocessing-free algorithm to generate two schemes for lossless data hiding in encrypted images. With either scheme, a processed image will be obtained after the encrypted image undergoes processing in the homomorphic encrypted domain. Besides retrieving a part of the hidden data without image decryption, the data hidden with the RES method can be extracted after decryption, even after some processing has been conducted on encrypted images. The experimental results show the efficacy and superior performance of the proposed schemes.

The existing image retrieval approaches focus on the behavior of a single user only in each query without considering the correlation of the behaviors of multiple users in performing similar queries. In fact, users would have similar behaviors while they have similar expectations during queries. Accordingly, this paper therefore proposes the interactive image retrieval framework with the Similar Behavior Learning model. The framework consists of two stages. In the first stage, the framework retrieves images with the content-based feature vector as preliminary query result for user selection. In the second stage, the SBL model determines the similarity of the user behavior and annotates label code to the selected images instantly. The images are indexed by label code can be retrieved more efficiently. Meanwhile, the selected images in preliminary result are used as additional information for retrieving better results at the end of the current query. Experiments show the promising results.

Recent studies of imbalanced data classification have shown that the imbalance ratio (IR) is not the only cause of performance loss in a classifier, as other data factors, such as small disjuncts, noise, and overlapping, can also make the problem difficult. The relationship between the IR and other data factors has been demonstrated, but to the best of our knowledge, there is no measurement of the extent to which class imbalance influences the classification performance of imbalanced data. In addition, it is also unknown which data factor serves as the main barrier for classification in a data set. In this article, we focus on the Bayes optimal classifier and examine the influence of class imbalance from a theoretical perspective. We propose an instance measure called the Individual Bayes Imbalance Impact Index (IBI3) and a data measure called the Bayes Imbalance Impact Index (BI3). IBI3 and BI3 reflect the extent of influence using only the imbalance factor, in terms of each minority class sample and the whole data set, respectively. Therefore, IBI3 can be used as an instance complexity measure of imbalance and BI3 as a criterion to demonstrate the degree to which imbalance deteriorates the classification of a data set. We can, therefore, use BI3 to access whether it is worth using imbalance recovery methods, such as sampling or cost-sensitive methods, to recover the performance loss of a classifier. The experiments show that IBI3 is highly consistent with the increase of the prediction score obtained by the imbalance recovery methods and that BI3 is highly consistent with the improvement in the F1 score obtained by the imbalance recovery methods on both synthetic and real benchmark data sets.

One of the most challenging problems in the field of online learning is concept drift, which deeply influences the classification stability of streaming data. If the data stream is imbalanced, it is even more difficult to detect concept drifts and make an online learner adapt to them. Ensemble algorithms have been found effective for the classification of streaming data with concept drift, whereby an individual classifier is built for each incoming data chunk and its associated weight is adjusted to manage the drift. However, it is difficult to adjust the weights to achieve a balance between the stability and adaptability of the ensemble classifiers. In addition, when the data stream is imbalanced, the use of a size-fixed chunk to build a single classifier can create further problems; the data chunk may contain too few or even no minority class samples (i.e., only majority class samples). A classifier built on such a chunk is unstable in the ensemble. In this article, we propose a chunk-based incremental learning method called adaptive chunk-based dynamic weighted majority (ACDWM) to deal with imbalanced streaming data containing concept drift. ACDWM utilizes an ensemble framework by dynamically weighting the individual classifiers according to their classification performance on the current data chunk. The chunk size is adaptively selected by statistical hypothesis tests to access whether the classifier built on the current data chunk is sufficiently stable. ACDWM has four advantages compared with the existing methods as follows: 1) it can maintain stability when processing nondrifted streams and rapidly adapt to the new concept; 2) it is entirely incremental, i.e., no previous data need to be stored; 3) it stores a limited number of classifiers to ensure high efficiency; and 4) it adaptively selects the chunk size in the concept drift environment. Experiments on both synthetic and real data sets containing concept drift show that ACDWM outperforms both state-of-the-art chunk-based and online methods.

Current unsupervised feature selection methods cannot well select the effective features from the corrupted data. To this end, we propose a robust unsupervised feature selection method under the robust principal component analysis (PCA) reconstruction criterion, which is named the adaptive weighted sparse PCA (AW-SPCA). In the proposed method, both the regularization term and the reconstruction error term are constrained by the $\ell _{2,1}$ -norm: the $\ell _{2,1}$ -norm regularization term plays a role in the feature selection, while the $\ell _{2,1}$ -norm reconstruction error term plays a role in the robust reconstruction. The proposed method is in a convex formulation, and the selected features by it can be used for robust reconstruction and clustering. Experimental results demonstrate that the proposed method can obtain better reconstruction and clustering performance, especially for the corrupted data.

Clustering ordinal data is a common task in data mining and machine learning fields. As a major type of categorical data, ordinal data is composed of attributes with naturally ordered possible values (also called categories interchangeably in this paper). However, due to the lack of dedicated distance metric, ordinal categories are usually treated as nominal ones, or coded as consecutive integers and treated as numerical ones. Both these two common ways will roughly define the distances between ordinal categories because the former way ignores the order relationship and the latter way simply assigns identical distances to different pairs of adjacent categories that may have intrinsically unequal distances. As a result, they may produce unsatisfactory ordinal data clustering results. This paper, therefore, proposes a novel ordinal data clustering algorithm, which iteratively learns: 1) The partition of ordinal dataset, and 2) the inter-category distances. To the best of our knowledge, this is the first attempt to dynamically adjust inter-category distances during the clustering process to search for a better partition of ordinal data. The proposed algorithm features superior clustering accuracy, low time complexity, fast convergence, and is parameter-free. Extensive experiments show its efficacy.

Linear discriminant analysis (LDA) is a classical supervised subspace learning technique that has wide applications. However, it is designed for vector only, which cannot exploit the tensor structures and may lead to suboptimal results when dealing with tensorial data. To address this problem, several multilinear LDA (MLDA) methods have been proposed to learn the subspaces from tensors. By exploiting the tensor structures, they achieve compact subspace representations, reduced parameter sizes, and improved robustness against the small sample size problem. However, existing MLDA methods do not take data uncertainty into account, fail to converge properly, or have to introduce additional tuning parameters for good convergence properties. In this paper, we therefore solve these limitations by proposing a probabilistic MLDA method for matrix inputs. Specifically, we propose a new generative model to incorporate structural information into the probabilistic framework, where each observed matrix is represented as a linear combination of collective and individual rank-one matrices. This provides our method with both the expressiveness of capturing discriminative features and nondiscriminative noise, and the capability of exploiting the 2-D tensor structures. To overcome the convergence problem of existing MLDAs, we develop an EM-type algorithm for parameter estimation, which has closed-form solutions with convergence guarantees. Experimental results on real-world datasets show the superiority of the proposed method to other probabilistic and MLDA variants.

Contrast enhancement (CE) of medical images is helpful to bring out the unclear content in the interested regions. Recently, reversible CE has been proposed so that the original version of a contrast-changed image can be exactly recovered. This property can be used to save storage space or facilitate the archiving system. To enhance the regions of interest (ROI) without introducing visual distortions, the technique of image segmentation (e.g. using Otsu's method) has been used to obtain the background before conducting the CE process. To segment the ROI more accurately, an interactive algorithm called GrabCut is employed in the proposed scheme. In addition, a new preprocessing strategy is adopted to preserve the image quality through the CE process. Consequently, the content in the selected regions can be better brought out while the reversibility of the CE process is achieved. The experimental results on 30 chest radiograph images and 20 magnetic resonance images have demonstrated the efficacy of the proposed scheme for reversible CE. The evaluation results are provided to show the better performances of the proposed method in achieving CE effects and preserving image quality.

Ordinal data are common in many data mining and machine learning tasks. Compared to nominal data, the possible values (also called categories interchangeably) of an ordinal attribute are naturally ordered. Nevertheless, since the data values are not quantitative, the distance between two categories of an ordinal attribute is generally not well defined, which surely has a serious impact on the result of the quantitative analysis if an inappropriate distance metric is utilized. From the practical perspective, ordinal-and-nominal-attribute categorical data, i.e., categorical data associated with a mixture of nominal and ordinal attributes, is common, but the distance metric for such data has yet to be well explored in the literature. In this paper, within the framework of clustering analysis, we therefore first propose an entropy-based distance metric for ordinal attributes, which exploits the underlying order information among categories of an ordinal attribute for the distance measurement. Then, we generalize this distance metric and propose a unified one accordingly, which is applicable to ordinal-and-nominal-attribute categorical data. Compared with the existing metrics proposed for categorical data, the proposed metric is simple to use and nonparametric. More importantly, it reasonably exploits the underlying order information of ordinal attributes and statistical information of nominal attributes for distance measurement. Extensive experiments show that the proposed metric outperforms the existing counterparts on both the real and benchmark data sets.

Hypervolume indicator based evolutionary algorithms have been reported to be very promising in many-objective optimization, but the high computational complexity of hypervolume calculation in high dimensions restrains its further applications and developments. In this paper, we develop a fast hypervolume approximation method with both improved speed and accuracy than the previous approximation methods via a new segmentation strategy. The proposed approach consists of two crucial process: segmentation and approximation. The segmentation process recursively finds areas easy to be measured and quantified from the original geometric figure as many as possible, and then divides the measurement of the rest areas into several subproblems. In the approximation process, an improved Monte Carlo simulation is developed to estimate these subproblems. Those two processes are mutually complementary to simultaneously improve the accuracy and the speed of hypervolume approximation. To validate its effectiveness, experimental studies on four widely-used instances are conducted and the simulation results show that the proposed method is ten times faster than other comparison algorithms with a same measurement error. Furthermore, we integrate an incremental version of this method into the framework of SMS-EMOA, and the performance of the integrated algorithm is also very competitive among the experimental algorithms.

Immune checkpoint inhibitors (ICI) have known associations with cardiotoxicity. However, a representative quantification of the adverse cardiovascular events and cardiovascular attendances amongst Asian users of ICI has been lacking. This retrospective cohort study identified all ICI users in Hong Kong, China, between 2013 and 2021. All patients were followed up until the end of 2021 for the primary outcome of major adverse cardiovascular event (MACE; a composite of cardiovascular mortality, myocardial infarction, heart failure, and stroke). Patients with prior diagnosis of any component of MACE were excluded from all MACE analyses. In total, 4324 patients were analyzed (2905 (67.2%) males; median age 63.5 years old (interquartile range 55.4-70.7 years old); median follow-up 1.0 year (interquartile range 0.4-2.3 years)), of whom 153 were excluded from MACE analyses due to prior events. MACE occurred in 116 (2.8%) with an incidence rate (IR) of 1.7 [95% confidence interval: 1.4, 2.0] events per 100 patient-years; IR was higher within the first year of follow-up (2.9 [2.3, 3.5] events per 100 patient-years). Cardiovascular hospitalization(s) occurred in 188 (4.4%) with 254 episodes (0.5% of all episodes) and 1555 days of hospitalization (1.3% of all hospitalized days), for whom the IR of cardiovascular hospitalization was 5.6 [4.6, 6.9] episodes per 100 person-years with 52.9 [39.8, 70.3] days' stay per 100 person-years. Amongst Asian users of ICI, MACE was uncommon, and a small proportion of hospitalizations were cardiovascular in nature. Most MACE and cardiovascular hospitalizations occurred during the first year after initiating ICI.

Decomposition-based evolutionary algorithms have shown great potential in many-objective optimization. However, the lack of theoretical studies on decomposition methods has hindered their further development and application. In this paper, we first theoretically prove that weight sum, Tchebycheff, and penalty boundary intersection decomposition methods are essentially interconnected. Inspired by this, we further show that highly customized dominance relationship can be derived from decomposition for any given decomposition vector. A new evolutionary algorithm is then proposed by applying the customized dominance relationship with adaptive strategy to each subpopulation of multiobjective to multiobjective framework. Experiments are conducted to compare the proposed algorithm with five state-of-the-art decomposition-based evolutionary algorithms on a set of well-known scaled many-objective test problems with 5 to 15 objectives. Simulation results have shown that the proposed algorithm can make better use of the decomposition vectors to achieve better performance. Further investigations on unscaled many-objective test problems verify the robust and generality of the proposed algorithm.

Existing convolutional neural networks (CNNs) have achieved significant performance on various real-life tasks, but a large number of parameters in convolutional layers requires huge storage and computation resources which makes it difficult to deploy CNNs on memory-constraint embedded devices. In this paper, we propose a novel compression method that generates the convolution filters in each layer by combining a set of learnable low-dimensional binary filter bases. The proposed method designs more compact convolution filters by stacking the linear combinations of these filter bases. Because of binary filters, the compact filters can be represented using less number of bits so that the network can be highly compressed. Furthermore, we explore the sparsity of coefficient through L1-ball projection when conducting linear combination to avoid overfitting. In addition, we analyze the compression performance of the proposed method in detail. Evaluations on four benchmark datasets under VGG-16 and ResNet-18 structures show that the proposed method can achieve a higher compression ratio with comparable accuracy compared with the existing state-of-the-art filter decomposition and network quantization methods.

Unsupervised cross-modal retrieval has received increasing attention recently, because of the extreme difficulty of labeling the explosive multimedia data. The core challenge of it is how to measure the similarities between multi-modal data without label information. In previous works, various distance metrics are selected for measuring the similarities and predicting whether samples belong to the same class. However, these predictions are not always right. Unfortunately, even a few wrong predictions can undermine the final retrieval performance. To address this problem, in this paper, we categorize predictions as solid and soft ones based on their confidence. We further categorize samples as solid and soft ones based on the predictions. We propose that these two kinds of predictions and samples should be treated differently. Besides, we find that the absolute values of similarities can represent not only the similarity but also the confidence of the predictions. Thus, we first design an elegant dot product fusion strategy to obtain effective inter-modal similarities. Subsequently, utilizing these similarities, we propose a generalized and flexible weighted loss function where larger weights are assigned to solid samples to increase the retrieval performance, and smaller weights are assigned to soft samples to decrease the disturbance of wrong predictions. Despite less information is used, empirical studies show that the proposed approach achieves the state-of-the-art retrieval performance.