Search

In this paper, we propose to tackle egocentric action recognition by suppressing background distractors and enhancing action-relevant interactions. The existing approaches usually utilize two independent branches to recognize egocentric actions, i.e., a verb branch and a noun branch. However, the mechanism to suppress distracting objects and exploit local human-object correlations is missing. To this end, we introduce two extra sources of information, i.e., the candidate objects' spatial location and their discriminative features, to enable concentration on the occurring interactions. We design a Symbiotic Attention withObject-centric featureAlignmentframework (SAOA) to provide meticulous reasoning between the actor and the environment. First, we introduce an object-centric feature alignment method to inject the local object features to the verb branch and noun branch. Second, we propose a symbiotic attention mechanism to encourage the mutual interaction between the two branches and select the most action-relevant candidates for classification. The framework benefits from the communication among the verb branch, the noun branch, and the local object information. Experiments based on different backbones and modalities demonstrate the effectiveness of our method. Notably, our framework achieves the state-of-the-art on the largest egocentric video dataset.

In most traditional machine learning algorithms, the training and testing datasets have identical distributions and feature spaces. However, these assumptions have not held in many real applications. Although transfer learning methods have been invented to fill this gap, they introduce new challenges as negative transfers (NTs). Most previous research considered NT a significant problem, but they pay less attention to solving it. This study will propose a transductive learning algorithm based on cellular learning automata (CLA) to alleviate the NT issue. Two famous learning automata (LA) entitled estimators are applied as estimator CLA in the proposed algorithms. A couple of new decision criteria called merit and and attitude parameters are introduced to CLA to limit NT. The proposed algorithms are applied to standard LA environments. The experiments show that the proposed algorithm leads to higher accuracy and less NT results.

Accurate object detection requires correct classification and high-quality localization. Currently, most of the single shot detectors (SSDs) conduct simultaneous classification and regression using a fully convolutional network. Despite high efficiency, this structure has some inappropriate designs for accurate object detection. The first one is the mismatch of bounding box classification, where the classification results of the default bounding boxes are improperly treated as the results of the regressed bounding boxes during the inference. The second one is that only one-time regression is not good enough for high-quality object localization. To solve the problem of classification mismatch, we propose a novel reg-offset-cls (ROC) module including three hierarchical steps: the regression of the default bounding box, the prediction of new feature sampling locations, and the classification of the regressed bounding box with more accurate features. For high-quality localization, we stack two ROC modules together. The input of the second ROC module is the output of the first ROC module. In addition, we inject a feature enhanced (FE) module between two stacked ROC modules to extract more contextual information. The experiments on three different datasets (i.e., MS COCO, PASCAL VOC, and UAVDT) are performed to demonstrate the effectiveness and superiority of our method. Without any bells or whistles, our proposed method outperforms state-of-the-art one-stage methods at a real-time speed. The source code is available at https://github.com/JialeCao001/HSD.

Model-based design is an important method of addressing problems associated with designing complex control systems. For complex dynamic systems in the presence of uncertainties, the modeling process from the first principles becomes extremely tedious and simplification in mechanism and parameter measurement may result in model inaccuracy. On the contrary, machine learning has the characteristic of fitting complicated equations, which makes it widely used in the research of model identification. However, it only brings a black-box model where the design schemes based on an analytical model cannot be applied. In this article, a simple and novel scheme for modeling and control of robotic manipulators is proposed; without prior knowledge, a dynamic model in an analytical form is obtained from artificially excited training data using the symbolic regression technique, and then, a controller is designed based on the dynamic model. Due to the ingenious experimental design, on one hand, the amount of training data is far less than the system identification method by machine learning. On the other hand, a decoupling feature is used in the model that greatly simplifies controller design. The experimental results on two-degree of freedom (DOF) and 6-DOF robotic manipulator simulators verify that the scheme is feasible and effective.

This article presents an analysis of data extraction for classification using correlation coefficient and fuzzy model. Several traditional methods of data extraction are used for classification that could not provide sufficient information for further step of data analysis on class. It needs refinement of features data to distinguish a class that differs from a traditional class. Thus, it proposes the feature tiny data (subfeature data) to find distinguish class from a traditional class using two methods such as correlation coefficient and fuzzy model to select features as well as subfeature for distinguishing class. In the first approach, the correlation coefficient methods with gradient descent technique are used to select features from the dataset and in the second approach, the fuzzy model with supreme of minimum value is considered to get subfeature data. As per the proposed model, some features (i.e., three features from the acoustic dataset, two features from the QCM dataset, and eight features from the audit dataset, etc.) and subfeatures (as per threshold value like 20 for acoustic; 10 for QCM, and 20 for audit, etc.) are selected based on correlation coefficient as well as fuzzy methods, respectively. Further, the probability approach is used to find the association and availability of subfeature data from the dimensional reduced database. The experimental results show the proposed framework identifies and selects both feature and subfeature data with the effectiveness of the new class. The comparison results of several classifiers on several datasets are explained in the experimental section.

Multivariate time series data (Mv-TSD) portray the evolving processes of the system(s) under examination in a multi-view manner. Factorization methods are salient for Mv-TSD analysis with the potentials of structural feature construction. However, research challenges remain in the derivation of factors due to scattered data distribution of Mv-TSD and intensive interferences/outliers embedded in the source data. The proposed Enhanced Bayesian Factorization approach (Enhanced-BF) addresses the challenges in three phases: (1) variant scale partitioning applies to Mv-TSD according to degree of amplitude and obtains the blocks of variant scales; (2) hierarchical Bayesian model for tensor factorization automatically derives the factors of each block with interferences suppressed; (3) Bayesian unification model merges those block factors to construct the final structural features. Enhanced-BF has been evaluated using a case study of brain data engineering with multivariate electroencephalogram (EEG). Experimental results indicate that the proposed method manifests robustness to the interferences and outperforms the counterparts in terms of operation efficiency and error when factorizing EEG tensor. Besides, Enhanced-BF excels in factorization-based analysis of ongoing autism spectrum disorder (ASD) EEG: 3 times speed-up in factorization and 87.35% accuracy in ASD discrimination. The latent factors can distinctly interpret the typical EEG characteristics of ASD subjects.

Heterogeneous domain adaptation (HDA) tackles the learning of cross-domain samples with both different probability distributions and feature representations. Most of the existing HDA studies focus on the single-source scenario. In reality, however, it is not uncommon to obtain samples from multiple heterogeneous domains. In this article, we study the multisource HDA problem and propose a conditional weighting adversarial network (CWAN) to address it. The proposed CWAN adversarially learns a feature transformer, a label classifier, and a domain discriminator. To quantify the importance of different source domains, CWAN introduces a sophisticated conditional weighting scheme to calculate the weights of the source domains according to the conditional distribution divergence between the source and target domains. Different from existing weighting schemes, the proposed conditional weighting scheme not only weights the source domains but also implicitly aligns the conditional distributions during the optimization process. Experimental results clearly demonstrate that the proposed CWAN performs much better than several state-of-the-art methods on four real-world datasets.

Being able to recognize emotional intensity is a desirable feature for a facial emotional recognition (FER) system. However, the development of such a feature is hindered by the paucity of intensity-labeled data for model training. To ameliorate the situation, the present study proposes using face morphing as a way of data augmentation to synthesize faces that express different degrees of a designated emotion. Such an approach has been successfully validated on humans and machines. Specifically, humans indeed perceived different levels of intensified emotions in these parametrically synthesized faces, and FER systems based on neural networks indeed showed improved sensitivities to intensities of different emotions when additionally trained on the synthesized faces. Overall, the proposed data augmentation method is not only simple and effective but also useful for building FER systems that recognize facial expressions of mixed emotions.

Facial micro-expression (ME) can disclose genuine and concealed human feelings. It makes MEs extensively useful in real-world applications pertaining to affective computing and psychology. Unfortunately, they are induced by subtle facial movements for a short duration of time, which makes the ME recognition, a highly challenging problem even for human beings. In automatic ME recognition, the well-known features encode either incomplete or redundant information, and there is a lack of sufficient training data. The proposed method, Micro-Expression Recognition by Analysing Spatial and Temporal Characteristics, MERASTC mitigates these issues for improving the ME recognition. It compactly encodes the subtle deformations using action units (AUs), landmarks, gaze, and appearance features of all the video frames while preserving most of the relevant ME information. Furthermore, it improves the efficacy by introducing a novel neutral face normalization for ME and initiating the utilization of gaze features in deep learning based ME recognition. The features are provided to the 2D convolutional neural network that jointly analyses the spatial and temporal behavior for correct ME classification. Experimental results on the publicly available datasets indicate that the proposed method exhibits better performance than the well-known methods.

Online users are typically active on multiple social media networks (SMNs), which constitute a multiplex social network. With improvements in cybersecurity awareness, users increasingly choose different usernames and provide different profiles on different SMNs. Thus, it is becoming increasingly challenging to determine whether given accounts on different SMNs belong to the same user; this can be expressed as an interlayer link prediction problem in a multiplex network. To address the challenge of predicting interlayer links, feature or structure information is leveraged. Existing methods that use network embedding techniques to address this problem focus on learning a mapping function to unify all nodes into a common latent representation space for prediction; positional relationships between unmatched nodes and their common matched neighbors (CMNs) are not utilized. Furthermore, the layers are often modeled as unweighted graphs, ignoring the strengths of the relationships between nodes. To address these limitations, we propose a framework based on multiple types of consistency between embedding vectors (MulCEVs). In MulCEV, the traditional embedding-based method is applied to obtain the degree of consistency between the vectors representing the unmatched nodes, and a proposed distance consistency index based on the positions of nodes in each latent space provides additional clues for prediction. By associating these two types of consistency, the effective information in the latent spaces is fully utilized. In addition, MulCEV models the layers as weighted graphs to obtain representation. In this way, the higher the strength of the relationship between nodes, the more similar their embedding vectors in the latent representation space will be. The results of our experiments on several real-world and synthetic datasets demonstrate that the proposed MulCEV framework markedly outperforms current embedding-based methods, especially when the number of training iterations is small.

Relation classification (RC) task is one of fundamental tasks of information extraction, aiming to detect the relation information between entity pairs in unstructured natural language text and generate structured data in the form of entity-relation triple. Although distant supervision methods can effectively alleviate the problem of lack of training data in supervised learning, they also introduce noise into the data, and still cannot fundamentally solve the long-tail distribution problem of the training instances. In order to enable the neural network to learn new knowledge through few instances like humans, this work focuses on few-shot relation classification (FSRC), where a classifier should generalize to new classes that have not been seen in the training set, given only a number of samples for each class. To make full use of the existing information and get a better feature representation for each instance, we propose to encode each class prototype in an adaptive way from two aspects. First, based on the prototypical networks, we propose an adaptive mixture mechanism to add label words to the representation of the class prototype, which, to the best of our knowledge, is the first attempt to integrate the label information into features of the support samples of each class so as to get more interactive class prototypes. Second, to more reasonably measure the distances between samples of each category, we introduce a loss function for joint representation learning to encode each support instance in an adaptive manner. Extensive experiments have been conducted on FewRel under different few-shot (FS) settings, and the results show that the proposed adaptive prototypical networks with label words and joint representation learning has not only achieved significant improvements in accuracy, but also increased the generalization ability of few-shot RC models.

Partial multilabel learning (PML) aims to learn from training data, where each instance is associated with a set of candidate labels, among which only a part is correct. The common strategy to deal with such a problem is disambiguation, that is, identifying the ground-truth labels from the given candidate labels. However, the existing PML approaches always focus on leveraging the instance relationship to disambiguate the given noisy label space, while the potentially useful information in label space is not effectively explored. Meanwhile, the existence of noise and outliers in training data also makes the disambiguation operation less reliable, which inevitably decreases the robustness of the learned model. In this article, we propose a prior label knowledge regularized self-representation PML approach, called PAKS, where the self-representation scheme and prior label knowledge are jointly incorporated into a unified framework. Specifically, we introduce a self-representation model with a low-rank constraint, which aims to learn the subspace representations of distinct instances and explore the high-order underlying correlation among different instances. Meanwhile, we incorporate prior label knowledge into the above self-representation model, where the prior label knowledge is regarded as the complement of features to obtain an accurate self-representation matrix. The core of PAKS is to take advantage of the data membership preference, which is derived from the prior label knowledge, to purify the discovered membership of the data and accordingly obtain more representative feature subspace for model induction. Enormous experiments on both synthetic and real-world datasets show that our proposed approach can achieve superior or comparable performance to state-of-the-art approaches.

Long-term visual place recognition (VPR) is challenging as the environment is subject to drastic appearance changes across different temporal resolutions, such as time of the day, month, and season. A wide variety of existing methods address the problem by means of feature disentangling or image style transfer but ignore the structural information that often remains stable even under environmental condition changes. To overcome this limitation, this article presents a novel structure-aware feature disentanglement network (SFDNet) based on knowledge transfer and adversarial learning. Explicitly, probabilistic knowledge transfer (PKT) is employed to transfer knowledge obtained from the Canny edge detector to the structure encoder. An appearance teacher module is then designed to ensure that the learning of appearance encoder does not only rely on metric learning. The generated content features with structural information are used to measure the similarity of images. We finally evaluate the proposed approach and compare it to state-of-the-art place recognition methods using six datasets with extreme environmental changes. Experimental results demonstrate the effectiveness and improvements achieved using the proposed framework. Source code and some trained models will be available at http://www.tianshu.org.cn.

Drug-drug interactions are one of the main concerns in drug discovery. Accurate prediction of drug-drug interactions plays a key role in increasing the efficiency of drug research and safety when multiple drugs are c o-prescribed. With various data sources that describe the relationships and properties between drugs, the comprehensive approach that integrates multiple data sources would be considerably effective in making high-accuracy prediction. In this paper, we propose a Deep Attention Neural Network based Drug-Drug Interaction prediction framework, abbreviated as DANN-DDI, to predict unobserved drug-drug interactions. First, we construct multiple drug feature networks and learn drug representations from these networks using the graph embedding method; then, we concatenate the learned drug embeddings and design an attention neural network to learn representations of drug-drug pairs; finally, we adopt a deep neural network to accurately predict drug-drug interactions. The experimental results demonstrate that our model DANN-DDI has improved prediction performance compared with state-of-the-art methods. Moreover, the proposed model can predict novel drug-drug interactions and drug-drug interaction-associated events.

This is the first work that incorporates recent advancements in "explainability" of machine learning (ML) to build a routing obfuscator called ObfusX. We adopt a recent metric---the SHAP value---which explains to what extent each layout feature can reveal each unknown connection for a recent ML-based split manufacturing attack model. The unique benefits of SHAP-based analysis include the ability to identify the best candidates for obfuscation, together with the dominant layout features which make them vulnerable. As a result, ObfusX can achieve better hit rate (97% lower) while perturbing significantly fewer nets when obfuscating using a via perturbation scheme, compared to prior work. When imposing the same wirelength limit using a wire lifting scheme, ObfusX performs significantly better in performance metrics (e.g., 2.4 times more reduction on average in percentage of netlist recovery).

Unsupervised domain adaptation methods have been proposed to tackle the problem of covariate shift by minimizing the distribution discrepancy between the feature embeddings of source domain and target domain. However, the standard evaluation protocols assume that the conditional label distributions of the two domains are invariant, which is usually not consistent with the real-world scenarios such as long-tailed distribution of visual categories. In this article, the imbalanced domain adaptation (IDA) is formulated for a more realistic scenario where both label shift and covariate shift occur between the two domains. Theoretically, when label shift exists, aligning the marginal distributions may result in negative transfer. Therefore, a novel cluster-level discrepancy minimization (CDM) is developed. CDM proposes cross-domain similarity learning to learn tight and discriminative clusters, which are utilized for both feature-level and distribution-level discrepancy minimization, palliating the negative effect of label shift during domain transfer. Theoretical justifications further demonstrate that CDM minimizes the target risk in a progressive manner. To corroborate the effectiveness of CDM, we propose two evaluation protocols according to the real-world situation and benchmark existing domain adaptation approaches. Extensive experiments demonstrate that negative transfer does occur due to label shift, while our approach achieves significant improvement on imbalanced datasets, including Office-31, Image-CLEF, and Office-Home.

Multi-view classification with limited sample size and data augmentation is a very common machine learning (ML) problem in medicine. With limited data, a triplet network approach for two-stage representation learning has been proposed. However, effective training and verifying the features from the representation network for their suitability in subsequent classifiers are still unsolved problems. Although typical distance-based metrics for the training capture the overall class separability of the features, the performance according to these metrics does not always lead to an optimal classification. Consequently, an exhaustive tuning with all feature-classifier combinations is required to search for the best end result. To overcome this challenge, we developed a novel nearest-neighbor (NN) validation strategy based on the triplet metric. This strategy is supported by a theoretical foundation to provide the best selection of the features with a lower bound of the highest end performance. The proposed strategy is a transparent approach to identify whether to improve the features or the classifier. This avoids the need for repeated tuning. Our evaluations on real-world medical imaging tasks (i.e., radiation therapy delivery error prediction and sarcoma survival prediction) show that our strategy is superior to other common deep representation learning baselines [i.e., autoencoder (AE) and softmax]. The strategy addresses the issue of feature's interpretability which enables more holistic feature creation such that the medical experts can focus on specifying relevant data as opposed to tedious feature engineering.

Most existing multiview clustering methods are based on the original feature space. However, the feature redundancy and noise in the original feature space limit their clustering performance. Aiming at addressing this problem, some multiview clustering methods learn the latent data representation linearly, while performance may decline if the relation between the latent data representation and the original data is nonlinear. The other methods which nonlinearly learn the latent data representation usually conduct the latent representation learning and clustering separately, resulting in that the latent data representation might be not well adapted to clustering. Furthermore, none of them model the intercluster relation and intracluster correlation of data points, which limits the quality of the learned latent data representation and therefore influences the clustering performance. To solve these problems, this article proposes a novel multiview clustering method via proximity learning in latent representation space, named multiview latent proximity learning (MLPL). For one thing, MLPL learns the latent data representation in a nonlinear manner which takes the intercluster relation and intracluster correlation into consideration simultaneously. For another, through conducting the latent representation learning and consensus proximity learning simultaneously, MLPL learns a consensus proximity matrix with k connected components to output the clustering result directly. Extensive experiments are conducted on seven real-world datasets to demonstrate the effectiveness and superiority of the MLPL method compared with the state-of-the-art multiview clustering methods.

This article dedicates to automatically explore efficient portrait parsing models that are easily deployed in edge computing or terminal devices. In the interest of the tradeoff between the resource cost and performance, we design the multiobjective reinforcement learning (RL)-based neural architecture search (NAS) scheme, which comprehensively balances the accuracy, parameters, FLOPs, and inference latency. Finally, under varying hyperparameter configurations, the search procedure emits a bunch of excellent objective-oriented architectures. The combination of two-stage training with precomputing and memory-resident feature maps effectively reduces the time consumption of the RL-based NAS method, so that we complete approximately 1000 search iterations in two GPU days. To accelerate the convergence of the lightweight candidate architecture, we incorporate knowledge distillation into the training of the search process. This also provides a reasonable evaluation signal to the RL controller that enables it to converge well. In the end, we conduct full training with outstanding Pareto-optimal architectures, so that a series of excellent portrait parsing models (with only approximately 0.3M parameters) is received. Furthermore, we directly transfer the architectures searched on CelebAMask-HQ (Portrait Parsing) to other portrait and face segmentation tasks. Finally, we achieve the state-of-the-art performance of 96.5% MIOU on EG1800 (portrait segmentation) and 91.6% overall F1-score on HELEN (face labeling). That is, our models significantly surpass the artificial network on the accuracy, but with lower resource consumption and higher real-time performance.

Recognizing the emotional state of a person within the image in real-world scenarios is a key problem in affective computing and has various promising applications. Local regions in the image, including different objects in the background scene and parts within the foreground body, usually have different contributions to emotion perception of the target person. This, however, has not been well exploited in most existing methods. In this paper, we propose to make relational region-level analysis to account for the different contributions of different regions to emotion recognition. For the background scene, we propose a Body-Object Attention (BOA) module to estimate the contributions of background objects to emotion recognition given the target foreground body. Within the foreground body, we propose a Body Part Attention (BPA) module to recalibrate the channel-wise body feature responses to attend on body parts that are more important. Moreover, we propose to model the emotion label dependency in real-world images, considering both the semantic meanings of these labels and their co-occurrence patterns. We evaluate the proposed method on the EMOTIC and CAER-S datasets, and experimental results show the superiority of our method compared with the state-of-the-art algorithms.

Learning from complementary labels (CLs) is a useful learning paradigm, where the CL specifies the classes that the instance does not belong to, instead of providing the ground truth as in the ordinary supervised learning scenario. In general, although it is less laborious and more efficient to collect CLs compared with ordinary labels, the less informative signal in the complementary supervision is less helpful to learn competent feature representation. Consequently, the final classifier's performance greatly deteriorates. In this article, we leverage generative adversarial networks (GANs) to derive an algorithm GAN-CL to effectively learn from CLs. In addition to the role in original GAN, the discriminator also serves as a normal classifier in GAN-CL, with the objective constructed partly with the complementary information. To further prove the effectiveness of our schema, we study the global optimality of both generator and discriminator for the GAN-CL under mild assumptions. We conduct extensive experiments on benchmark image datasets using deep models, to demonstrate the compelling improvements, compared with state-of-the-art CL learning approaches.

A common drawback of EEG-based emotion recognition is that volume conduction effects of the human head introduce interchannel dependence and result in highly correlated information among most EEG features. These highly correlated EEG features cannot provide extra useful information, and they actually reduce the performance of emotion recognition. However, the existing feature selection methods, commonly used to remove redundant EEG features for emotion recognition, ignore the correlation between the EEG features or utilize a greedy strategy to evaluate the interdependence, which leads to the algorithms retaining the correlated and redundant features with similar feature scores in the EEG feature subset. To solve this problem, we propose a novel EEG feature selection method for emotion recognition, termed global redundancy minimization in orthogonal regression (GRMOR). GRMOR can effectively evaluate the dependence among all EEG features from a global view and then select a discriminative and nonredundant EEG feature subset for emotion recognition. To verify the performance of GRMOR, we utilized three EEG emotional data sets (DEAP, SEED, and HDED) with different numbers of channels (32, 62, and 128). The experimental results demonstrate that GRMOR is a promising tool for redundant feature removal and informative feature selection from highly correlated EEG features.

Zero-Shot Learning (ZSL) aims to recognize unseen classes that never appear during training. Recently, generative adversarial networks (GANs) have been introduced to convert ZSL into a supervised learning problem by synthesizing unseen visual features. However, since unseen classes are never experienced for the generator during training, the synthesized unseen visual features often become heavily biased towards seen classes, or sometimes there is even no meaningful class that can be assigned to them. This is known as the \textit{bias problem}. In this paper, we propose a novel method, namely Adaptive Bias-Aware GAN (ABA-GAN), to alleviate generating biased visual features. For this purpose, we build a semantic adversarial network to regularize the feature generator. Specifically, an adaptive adversarial loss is proposed to constrain the feature distributions, which avoids the generation of meaningless visual features. Meanwhile, a domain divider is presented to explicitly distinguish synthesized visual features between seen and unseen domains, such that the bias towards seen classes can be alleviated. Moreover, we propose a novel metric named bias score (BS) to explicitly quantify the degree of the strong bias. Extensive experiments on four widely used benchmark datasets demonstrate that our proposed method outperforms the state-of-the-art approaches under both ZSL and GZSL protocols.

Domain generalization aims to generalize a network trained on multiple domains to unknown yet related domains. Operating under the assumption that invariant information generalizes well to unknown domains, previous work has aimed to minimize the discrepancies amongst distributions across given domains. However, without prior regularization of feature distributions in unknown domains, the network in practice overfits the invariant information in the given domains. Moreover, if there are insufficient samples in given domains, then domain generalizability is limited, as diverse domain variations are not captured. To address these two drawbacks, we propose to explicitly map features in known and unknown domains onto latent space in a fixed Gaussian mixture distribution by variational coding. As a result, features in different classes follow Gaussian distributions with different mean values. The predefined latent space narrows discrepancies between known and unknown domains and effectively separates samples into different classes. Moreover, we propose to perturb sample features with gradients from the distribution regularized loss. This perturbation generates samples beyond but near the latent space of prior distributions, which has a profound impact on domain variations. Experiments and visualizations demonstrate the effectiveness of our proposed method.

This article proposes a robust end-to-end deep learning-induced fault recognition scheme by stacking multiple sparse-denoising autoencoders with a Softmax classifier, called stacked spare-denoising autoencoder (SSDAE)-Softmax, for the fault identification of complex industrial processes (CIPs). Specifically, sparse denoising autoencoder (SDAE) is established by integrating a sparse AE (SAE) with a denoising AE (DAE) for the low-dimensional but intrinsic feature representation of the CIP monitoring data (CIPMD) with possible noise contamination. SSDAE-Softmax is established by stacking multiple SDAEs with a layerwise pretraining procedure, and a Softmax classifier with a global fine-tuning strategy. Furthermore, SSDAE-Softmax hyperparameters are optimized by a relatively new global optimization algorithm, referred to as the state transition algorithm (STA). Benefiting from the deep learning-based feature representation scheme with the STA-based hyperparameter optimization, the underlying intrinsic characteristics of CIPMD can be learned automatically and adaptively for accurate fault identification. A numeric simulation system, the benchmark Tennessee Eastman process (TEP), and a real industrial process, that is, the continuous casting process (CCP) from a top steel plant of China, are used to validate the performance of the proposed method. Experimental results show that the proposed SSDAE-Softmax model can effectively identify various process faults, and has stronger robustness and adaptability against the noise interference in CIPMD for the process monitoring of CIPs.

Capturing the dynamics of facial expression progression in video is an essential and challenging task for facial expression recognition (FER). In this paper, we propose an effective framework to address this challenge. We develop a C3D-based network architecture, 3D-Inception-ResNet, to extract spatial-temporal features from the dynamic facial expression image sequence. A Spatial-Temporal and Channel Attention Module (STCAM) is proposed to explicitly exploit the holistic spatial-temporal and channel-wise correlations among the extracted features. Specifically, the proposed STCAM calculates a channel-wise and a spatial-temporal-wise attention map to enhance the features along the corresponding feature dimensions for more representative features. We evaluate our method on three popular dynamic facial expression recognition datasets, CK+, Oulu-CASIA and MMI. Experimental results show that our method achieves better or comparable performance compared to the state-of-the-art approaches.

Alzheimer's disease (AD) is a neurodegenerative disease of the human brain that affects neurotransmitters, tissue, and neurons that impair the senses, memories, and behaviors. Still, now there is no remedy for Alzheimer's disease. Even so, prescribed drugs can help reduce the development of the disease. That's why Alzheimer's early detection is very essential for treatment, and further research. Very limited numbers of trained samples and the higher volume of feature descriptions are the major difficulties in early diagnosis of Alzheimer's disease using different classification strategies. In this article, we proposed and related Alzheimer's disease early diagnostic method using Mild Cognitive Impairment (MCI), Structural Magnetic Resonance (sMR) imaging for AD-discrimination and healthy control participants (HC) with Import Vector Machine (IVM), Regularized Extreme Learning Machine (RELM) and a Support vector machine (SVM).The greedy score-based strategy for choosing essential function vectors is used. Furthermore, a discriminatory, kernel-based method is taken to treat dynamic data transformations. For volume sMR scan image data from Alzheimer's disease neuroimaging initiative (ADNI) repositories, we compare the performance of these classification models. An ADNI datasets experimental study reveals that RELM can greatly enhance the accuracy for classification of AD from MCIs as well as HC individuals along with feature selection methodology.

Clustering algorithms based on deep neural networks have been widely studied for image analysis. Most existing methods require partial knowledge of the true labels, namely, the number of clusters, which is usually not available in practice. In this article, we propose a Bayesian nonparametric framework, deep nonparametric Bayes (DNB), for jointly learning image clusters and deep representations in a doubly unsupervised manner. In doubly unsupervised learning, we are dealing with the problem of ``unknown unknowns,'' where we estimate not only the unknown image labels but also the unknown number of labels as well. The proposed algorithm alternates between generating a potentially unbounded number of clusters in the forward pass and learning the deep networks in the backward pass. With the help of the Dirichlet process mixtures, the proposed method is able to partition the latent representations space without specifying the number of clusters a priori. An important feature of this work is that all the estimation is realized with an end-to-end solution, which is very different from the methods that rely on post hoc analysis to select the number of clusters. Another key idea in this article is to provide a principled solution to the problem of ``trivial solution'' for deep clustering, which has not been much studied in the current literature. With extensive experiments on benchmark datasets, we show that our doubly unsupervised method achieves good clustering performance and outperforms many other unsupervised image clustering methods.

This paper develops parameter instability and structural change tests within predictive regressions for economic systems governed by persistent vector autoregressive dynamics. Specifically, in a setting where all – or a subset – of the variables may be fractionally integrated and the predictive relation may feature cointegration, we provide sup-Wald break tests that are constructed using the Local speCtruM (LCM) approach. The new tests cover both parameter variation and multiple structural changes with unknown break dates, and the number of breaks being known or unknown. We establish asymptotic limit theory for the tests, showing that it coincides with standard testing procedures. As a consequence, existing critical values for tied-down Bessel processes may be applied, without modification. We implement the new structural change tests to explore the stability of the fractionally cointegrating relation between implied- and realized volatility (IV and RV). Moreover, we assess the relative efficiency of IV forecasts against a challenging time-series benchmark constructed from high-frequency data. Unlike existing studies, we find evidence that the IV–RV cointegrating relation is unstable, and that carefully constructed time-series forecasts are more efficient than IV in capturing low-frequency movements in RV.

Available data in machine learning applications is becoming increasingly complex, due to higher dimensionality and difficult classes. There exists a wide variety of approaches to measuring complexity of labeled data, according to class overlap, separability or boundary shapes, as well as group morphology. Many techniques can transform the data in order to find better features, but few focus on specifically reducing data complexity. Most data transformation methods mainly treat the dimensionality aspect, leaving aside the available information within class labels which can be useful when classes are somehow complex. This paper proposes an autoencoder-based approach to complexity reduction, using class labels in order to inform the loss function about the adequacy of the generated variables. This leads to three different new feature learners, Scorer, Skaler and Slicer. They are based on Fisher's discriminant ratio, the Kullback-Leibler divergence and least-squares support vector machines, respectively. They can be applied as a preprocessing stage for a binary classification problem. A thorough experimentation across a collection of 27 datasets and a range of complexity and classification metrics shows that class-informed autoencoders perform better than 4 other popular unsupervised feature extraction techniques, especially when the final objective is using the data for a classification task., This paper has been accepted for publication by IEEE Transactions on Pattern Analysis and Machine Intelligence

Vibration signals are widely utilized in many fields, which can reflect machine health state. Those typical deep learning techniques cannot learn impulsive features from vibration signals due to interference of strong background noise. Supervised learning techniques greatly rely on vast labeled data, which limits implementation of deep learning in industry applications. Hence, a new deep neural network (DNN), sparse representation convolutional autoencoder (SRCAE) is proposed to extract impulsive components of vibration signals for machinery fault diagnosis in an unsupervised manner. A sparse representation (SR) block is proposed to extract impulsive components of vibration signals and transform time-domain signal to a sparse domain by sparse mapping of convolutional graph. The SR block is inserted into deep network to remove noise and learn impulsive features for machinery fault diagnosis. Furthermore, an unsupervised selective feature transmission mechanism (USFTM) is proposed to improve training efficiency and realize feature filtering simultaneously. Finally, the effectiveness of SRCAE is verified on rotary machine fault diagnosis experiments. The testing results show that SRCAE has good noise filtering and impulsive components extraction performance. The recognition accuracy of SRCAE reached 97.16% based on the five-fold cross validation, which demonstrates the outperformance of SRCAE in comparison with state-of-the-art DNNs.

Domain adaptation (DA) refers to generalize a learning technique across the source domain and target domain under different distributions. Therefore, the essential problem in DA is how to reduce the distribution discrepancy between the source and target domains. Typical methods are to embed the adversarial learning technique into deep networks to learn transferable feature representations. However, existing adversarial related DA methods may not sufficiently minimize the distribution discrepancy. In this paper, a domain adaptation method minimum adversarial distribution discrepancy (MADD) is proposed by combining feature distribution with adversarial learning. Specifically, we design a novel divergence metric loss, named maximum mean discrepancy based on conditional entropy (MMD-CE), and embed it in the adversarial domain adaptation network. The proposed MMD-CE loss can address two problems: 1) the misalignment from different class distributions between domains, 2) the equilibrium challenge issue in adversarial domain adaptation. Comparative experiments on Office-31, ImageCLEF-DA, and Office-Home datasets with state-of-the-art methods show that, our method has some advantageous performances.

Broad learning system (BLS) is a novel and efficient model, which facilitates representation learning and classification by concatenating feature nodes and enhancement nodes. In spite of the efficient properties, BLS is still suboptimal when facing with imbalance problem. Besides, outliers and noises in imbalanced data remain a challenge for BLS. To address the above issues, in this paper we firstly propose a weighted BLS, which assigns a weight to each training sample, and adopt a general weighting scheme, which augments the weight of samples from the minority class. To further explore the prior distribution of original data, we design a density based weight generation mechanism to guide the specific weight matrix generation and propose the adaptive weighted broad learning system (AWBLS). This mechanism considers the inter-class and intra-class distance simultaneously in the density calculation. Finally, we propose the incremental weighted ensemble broad learning system (IWEB) by utilizing a progressive mechanism to further improve the stability and robustness of AWBLS. Extensive comparative experiments on 30 real-world data sets verfy that IWEB outperforms most of the imbalance ensemble classification methods.

Depth maps acquired with ToF cameras have a limited accuracy due to the high noise level and to the multi-path interference. Deep networks can be used for refining ToF depth, but their training requires real world acquisitions with ground truth, which is complex and expensive to collect. A possible workaround is to train networks on synthetic data, but the domain shift between the real and synthetic data reduces the performances. In this paper, we propose three approaches to perform unsupervised domain adaptation of a depth denoising network from synthetic to real data. These approaches are respectively acting at the input, at the feature and at the output level of the network. The first approach uses domain translation networks to transform labeled synthetic ToF data into a representation closer to real data, that is then used to train the denoiser. The second approach tries to align the network internal features related to synthetic and real data. The third approach uses an adversarial loss, implemented with a discriminator trained to recognize the ground truth statistic, to train the denoiser on unlabeled real data. Experimental results show that the considered approaches are able to outperform other state-of-the-art techniques and achieve superior denoising performances.

In this article, the unsupervised domain adaptation problem, where an approximate inference model is to be learned from a labeled dataset and expected to generalize well on an unlabeled dataset, is considered. Unlike the existing work, we explicitly unveil the importance of the latent variables produced by the feature extractor, that is, encoder, where contains the most representative information about their input samples, for the knowledge transfer. We argue that an estimator of the representation of the two datasets can be used as an agent for knowledge transfer. To be specific, a novel variational inference approach is proposed to approximate a latent distribution from the unlabeled dataset that can be used to accurately predict its input samples. It is demonstrated that the discriminative knowledge of the latent distribution that is learned from the labeled dataset can be progressively transferred to that is learned from the unlabeled dataset by simultaneously optimizing the estimator via the variational inference and our proposed regularization for shifting the mean of the estimator. The experiments on several benchmark datasets demonstrate that the proposed method consistently outperforms state-of-the-art methods for both object classification and digit classification.

CNN-based salient object detection (SOD) methods achieve impressive performance. However, the way semantic information is encoded in them and whether they are category-agnostic is less explored. One major obstacle in studying these questions is the fact that SOD models are built on top of the ImageNet pre-trained backbones which may cause information leakage and feature redundancy. To remedy this, here we first propose an extremely light-weight holistic model tied to the SOD task that can be freed from classification backbones and trained from scratch, and then employ it to study the semantics of SOD models. With the holistic network and representation redundancy reduction by a novel dynamic weight decay scheme, our model has only 100K parameters, ∼ 0.2% of parameters of large models, and performs on par with SOTA on popular SOD benchmarks. Using CSNet, we find that a) SOD and classification methods use different mechanisms, b) SOD models are category insensitive, c) ImageNet pre-training is not necessary for SOD training, and d) SOD models require far fewer parameters than the classification models. The source code is publicly available at https://mmcheng.net/sod100k/.

Feature selection has attracted a lot of attention in obtaining discriminative and non-redundant features from high-dimension data. Compared with traditional filter and wrapper methods, embedded methods can obtain a more informative feature subset by fully considering the importance of features in the classification tasks. However, the existing embedded methods emphasize the above importance of features and mostly ignore the correlation between the features, which leads to retain the correlated and redundant features with similar scores in the feature subset. To solve the problem, we propose a novel supervised embedded feature selection framework, called feature selection under global redundancy minimization in orthogonal regression (GRMOR). The proposed framework can effectively recognize redundant features from a global view of redundancy among the features. We also incorporate the large margin constraint into GRMOR for robust multi-class classification. Compared with the traditional embedded methods based on least square regression, the proposed framework utilizes orthogonal regression to preserve more discriminative information in the subspace, which can help accurately rank the importance of features in the classification tasks. Experimental results on twelve public datasets demonstrate that the proposed framework can obtain superior classification performance and redundancy removal performance than twelve other feature selection methods.

Despite that pre-trained word embedding models have advanced a wide range of natural language processing applications, they ignore the contextual information and meaning within the text. In this paper, we investigate the potential of the pre-trained Arabic BERT (Bidirectional Encoder Representations from Transformers) model to learn universal contextualized sentence representations aiming to showcase its usefulness for Arabic text Multi-class categorization. We propose to exploit the pre-trained AraBERT for contextual text representation learning in two different ways, transfer learning model and feature extractor. On the one hand, we employ the Arabic BERT (AraBERT) model after fine-tuning its parameters on the OSAC datasets to transfer its knowledge for the Arabic text categorization. On the other hand, we inquire into AraBERT performance, as a feature extractor model, by combining it with several classifiers, including CNN, LSTM, Bi-LSTM, MLP, and SVM. Finally, we conduct an exhaustive set of experiments comparing two BERT models, namely AraBERT and multilingual BERT. The findings show that the fine-tuned AraBERT model accomplishes state-of-the-art performance results and attains up to 99% in terms of F1-score and accuracy.

The proposed work aims to solve data sparsity problem in the recommendation system. It handles two-level pre-processing techniques to reduce the data size at the item level. Additional resources like items genre, tag, and time are added to learn and analyse the behaviour of the user preferences in-depth. The advantage of the proposed method is to recommend the item, based on user interest pattern and avoid recommending the outdated items. User information are grouped based on similar item genre and tag feature. This effectively handle overlapping conditions that exist on item’s genre, as it has more than one genre at initial level. Further, based on time, it analyses the user non-static interest. Overall it reduces the dimensions which is an initial way to prepare data, to analyse hidden pattern. To enhance the performance, the proposed method utilized Apache’s spark Mllib FP-Growth and association rule mining approach in a distributed environment. To reduce the computation cost of constructing tree in FP-Growth, the candidate data set is stored in matrix form. The experiments were conducted using MovieLens data set. The observed results shows that the proposed method achieves 4% increase in accuracy when compared to earlier methods.

This article addresses the problem of consistently identifying a single transfer function in a network of dynamic systems using only observational data. It is assumed that the topology is partially known, the forcing inputs are not measured, and that only a subset of the nodes outputs is accessible. The developed technique is applicable to scenarios encompassing confounding variables and feedback loops, which are complicating factors potentially introducing bias in the estimate of the transfer function. The results are based on the prediction of the output node using the input node along with a set of additional auxiliary variables which are selected only from the observed nodes. Similar prediction error methods provide only sufficient conditions for the appropriate choice of auxiliary variables and assume a priori information about the location of strictly causal operators in the network. In this article, such an a priori knowledge is not required. A most remarkable feature of our approach is that the conditions for the selection of the auxiliary variables are purely graphical. Furthermore, within single-output prediction methods such conditions are proven to

Few-shot learning, aiming to learn novel concepts from one or a few labeled examples, is an interesting and very challenging problem with many practical advantages. Existing few-shot methods usually utilize data of the same classes to train the feature embedding module and in a row, which is unable to learn adapting to new tasks. Besides, traditional few-shot models fail to take advantage of the valuable relations of the support-query pairs, leading to performance degradation. In this article, we propose a transductive relation-propagation graph neural network (GNN) with a decoupling training strategy (TRPN-D) to explicitly model and propagate such relations across support-query pairs, and empower the few-shot module the ability of transferring past knowledge to new tasks via the decoupling training. Our few-shot module, namely TRPN, treats the relation of each support-query pair as a graph node, named relational node, and resorts to the known relations between support samples, including both intraclass commonality and interclass uniqueness. Through relation propagation, the model could generate the discriminative relation embeddings for support-query pairs. To the best of our knowledge, this is the first work that decouples the training of the embedding network and the few-shot graph module with different tasks, which might offer a new way to solve the few-shot learning problem. Extensive experiments conducted on several benchmark datasets demonstrate that our method can significantly outperform a variety of state-of-the-art few-shot learning methods.

In the context of streaming data, learning algorithms often need to confront several unique challenges, such as concept drift, label scarcity, and high dimensionality. Several concept drift-aware data stream learning algorithms have been proposed to tackle these issues over the past decades. However, most existing algorithms utilize a supervised learning framework and require all true class labels to update their models. Unfortunately, in the streaming environment, requiring all labels is unfeasible and not realistic in many real-world applications. Therefore, learning data streams with minimal labels is a more practical scenario. Considering the problem of the curse of dimensionality and label scarcity, in this article, we present a new semisupervised learning technique for streaming data. To cure the curse of dimensionality, we employ a denoising autoencoder to transform the high-dimensional feature space into a reduced, compact, and more informative feature representation. Furthermore, we use a cluster-and-label technique to reduce the dependency on true class labels. We employ a synchronization-based dynamic clustering technique to summarize the streaming data into a set of dynamic microclusters that are further used for classification. In addition, we employ a disagreement-based learning method to cope with concept drift. Extensive experiments performed on many real-world datasets demonstrate the superior performance of the proposed method compared to several state-of-the-art methods.

Deep neural networks have achieved breakthrough improvement in various application fields. Nevertheless, they usually suffer from a time-consuming training process because of the complicated structures of neural networks with a huge number of parameters. As an alternative, a fast and efficient discriminative broad learning system (BLS) is proposed, which takes the advantages of flat structure and incremental learning. The BLS has achieved outstanding performance in classification and regression problems. However, the previous studies ignored the reason why the BLS can generalize well. In this article, we focus on the interpretation from the viewpoint of the frequency domain. We discover the existence of the frequency principle in BLS, i.e., the BLS preferentially captures low-frequency components quickly and then fits the high frequencies during the incremental process of adding feature nodes and enhancement nodes. The frequency principle may be of great inspiration for expanding the application of BLS.

Traffic forecasting has attracted considerable attention due to its importance in proactive urban traffic control and management. Scholars and engineers have exerted considerable efforts in improving the performance of traffic forecasting algorithms in terms of accuracy, reliability, and efficiency. Spatial feature representation of traffic flow is a core component that greatly influences traffic forecasting performance. In previous studies, several spatial attributes of traffic flow are ignored due to the following issues: a) traffic flow propagation does not comply with the road network, b) the spatial pattern of traffic flow varies over time, and c) single adjacent matrix cannot handle the complex and hierarchical urban traffic flow. To address the abovementioned issues, this study proposes a novel traffic forecasting algorithm called traffic transformer, which achieves great success in natural language processing. The multihead attention mechanism and stacking layers enable the transformer to learn dynamic and hierarchical features in sequential data. Two components, namely, global encoder and global-local decoder, are proposed to extract and fuse the spatial patterns globally and locally. Experimental results indicate that the proposed traffic transformer outperforms state-of-the-art methods. The learned dynamic and hierarchical features of traffic flow can help achieve a better understanding of spatial dependency of traffic flow for effective and efficient traffic control and management strategies.

Domain adaptation aims at extracting knowledge from auxiliary source domains to assist the learning task in a target domain. In classification problems, since the distributions of the source and target domains are different, directly using source data to build a classifier for the target domain may hamper the classification performance on the target data. Fortunately, in many tasks, there can be some features that are transferable, i.e., the source and target domains share similar properties. On the other hand, it is common that the source data contain noisy features which may degrade the learning performance in the target domain. This issue, however, is barely studied in existing works. In this paper, we propose to find a feature subset that is transferable across the source and target domains. As a result, the domain discrepancy measured on the selected features can be reduced. Moreover, we seek to find the most discriminative features for classification. To achieve the above goals, we formulate a new sparse learning model that is able to jointly reduce the domain discrepancy and select informative features for classification. We develop two optimization algorithms to address the derived learning problem. Extensive experiments on real-world data sets demonstrate the effectiveness of the proposed method.

Facial microexpressions offer useful insights into subtle human emotions. This unpremeditated emotional leakage exhibits the true emotions of a person. However, the minute temporal changes in the video sequences are very difficult to model for accurate classification. In this article, we propose a novel spatiotemporal architecture search algorithm, AutoMER for microexpression recognition (MER). Our main contribution is a new parallelogram design-based search space for efficient architecture search. We introduce a spatiotemporal feature module named 3-D singleton convolution for cell-level analysis. Furthermore, we present four such candidate operators and two 3-D dilated convolution operators to encode the raw video sequences in an end-to-end manner. To the best of our knowledge, this is the first attempt to discover 3-D convolutional neural network (CNN) architectures with a network-level search for MER. The searched models using the proposed AutoMER algorithm are evaluated over five microexpression data sets: CASME-I, SMIC, CASME-II, CAS(ME) ∧2 , and SAMM. The proposed generated models quantitatively outperform the existing state-of-the-art approaches. The AutoMER is further validated with different configurations, such as downsampling rate factor, multiscale singleton 3-D convolution, parallelogram, and multiscale kernels. Overall, five ablation experiments were conducted to analyze the operational insights of the proposed AutoMER.

Spatial downsampling layers are favored in convolutional neural networks (CNNs) to downscale feature maps for larger receptive fields and less memory consumption. However, for discriminative tasks, there is a possibility that these layers lose the discriminative details due to improper pooling strategies, which could hinder the learning process and eventually result in suboptimal models. In this paper, we present a unified framework over the existing downsampling layers (e.g., average pooling, max pooling, and strided convolution) from a local importance view. In this framework, we analyze the issues of these widely-used pooling layers and figure out the criteria for designing an effective downsampling layer. According to this analysis, we propose a conceptually simple, general, and effective pooling layer based on local importance modeling, termed as {\em Local Importance-based Pooling} (LIP). LIP can automatically enhance discriminative features during the downsampling procedure by learning adaptive importance weights based on inputs. Experiment results show that LIP consistently yields notable gains with different depths and different architectures on ImageNet classification. In the challenging MS COCO dataset, detectors with our LIP-ResNets as backbones obtain a consistent improvement ($\ge 1.4\%$) over the vanilla ResNets, and especially achieve the current state-of-the-art performance in detecting small objects under the single-scale testing scheme., Comment: ICCV 2019 camera ready version. Code: https://github.com/sebgao/LIP

Deep learning greatly empowers Inertial Measurement Unit (IMU) sensors for various mobile sensing applications, including human activity recognition, human-computer interaction, localization and tracking, and many more. Most existing works require substantial amounts of well-curated labeled data to train IMU-based sensing models, which incurs high annotation and training costs. Compared with labeled data, unlabeled IMU data are abundant and easily accessible. In this work, we present LIMU-BERT, a novel representation learning model that can make use of unlabeled IMU data and extract generalized rather than task-specific features. LIMU-BERT adopts the principle of self-supervised training of the natural language model BERT to effectively capture temporal relations and feature distributions in IMU sensor measurements. However, the original BERT is not adaptive to mobile IMU data. By meticulously observing the characteristics of IMU sensors, we propose a series of techniques and accordingly adapt LIMU-BERT to IMU sensing tasks. The designed models are lightweight and easily deployable on mobile devices. With the representations learned via LIMU-BERT, task-specific models trained with limited labeled samples can achieve superior performances. We extensively evaluate LIMU-BERT with four open datasets. The results show that the LIMU-BERT enhanced models significantly outperform existing approaches in two typical IMU sensing applications.

With the memory-resource-limited constraints, class-incremental learning (CIL) usually suffers from the ``catastrophic forgetting'' problem when updating the joint classification model on the arrival of newly added classes. To cope with the forgetting problem, many CIL methods transfer the knowledge of old classes by preserving some exemplar samples into the size-constrained memory buffer. To utilize the memory buffer more efficiently, we propose to keep more auxiliary low-fidelity exemplar samples, rather than the original real-high-fidelity exemplar samples. Such a memory-efficient exemplar preserving scheme makes the old-class knowledge transfer more effective. However, the low-fidelity exemplar samples are often distributed in a different domain away from that of the original exemplar samples, that is, a domain shift. To alleviate this problem, we propose a duplet learning scheme that seeks to construct domain-compatible feature extractors and classifiers, which greatly narrows down the above domain gap. As a result, these low-fidelity auxiliary exemplar samples have the ability to moderately replace the original exemplar samples with a lower memory cost. In addition, we present a robust classifier adaptation scheme, which further refines the biased classifier (learned with the samples containing distillation label knowledge about old classes) with the help of the samples of pure true class labels. Experimental results demonstrate the effectiveness of this work against the state-of-the-art approaches. We will release the code, baselines, and training statistics for all models to facilitate future research.

Background and objective Heart murmur characterization is a crucial part of cardiac auscultation for determining the potential etiology and severity of heart diseases. One such helpful murmur characterization is the sonic qualities, which reflect both structural and hemodynamical states of the heart. Therefore, the objective is to develop a machine learning based solution for classifying murmur qualities. Methods Four medically defined murmur qualities, namely the musical quality, blowing-like quality, coarse quality, and soft quality were examined. Feature was extracted from heart murmurs signals in their time domain, frequency domain, time-frequency domain, and phase space domain. Sequential forward floating selection (SFFS) was implemented along with three classifiers, including k-nearest neighbor (KNN), Naive-Bayes (NB), and linear support vector machine (SVM). Results It was found that multi-domain features are suited for better classification results and linear SVM was able to achieve a better balance between performance and the size of feature subsets among tested classifiers. Using the derived features, classification accuracies of 86%, 91%, 90%, and 84% were achieved for musical quality, blowing-like quality, coarse quality, and soft quality classifications respectively. Conclusions The study demonstrated that it is possible to effectively characterize heart murmur through its diagnostic characteristics instead of drawing direct conclusions, which is helpful for retaining versatility and generality found in the conventional cardiac auscultation.