Search

Weakly-supervised anomaly detection (WSAD) has gained increasing attention due to its core idea of enhancing the performance of unsupervised anomaly detection by leveraging prior knowledge from a limited number of labeled anomalies. In this paper, we introduce a novel WSAD framework that surpasses current state-of-the-art methods in terms of accuracy, exhibits greater robustness to data uncertainty, and is more efficient in utilizing limited labeled anomalies. Our method is built upon twin fuzzy networks (TFN) that learn robust fuzzy if-then rules from a pairwise training set. TFN can extract informative prototypes of training instances, exploiting the very few labeled anomalies efficiently. A two-stage sequential training scheme, comprising fuzzy C-means clustering and interpolation consistency regularization, ensures that the fuzzy rules form a solid foundation for anomaly detection while improving TFN's generalization ability. The training process of TFN relies on closed-form optimization rather than gradient-based methods, leading to significantly faster training speeds. Comprehensive experiments conducted on numerous real-world datasets confirm the advantages of the TFN framework over existing alternatives.

Compared to traditional centralized clustering, distributed clustering offers the advantage of parallel processing of data from different sites, enhancing the efficiency of clustering while preserving the privacy of the data at each site. However, most existing distributed clustering techniques require manual tuning of several parameters or hyperparameters, which can pose challenges for practical applications. This paper introduces a novel parameter-free distributed clustering framework known as distributed torque clustering (DTC). When dealing with data or subdata distributed across various sites, DTC predominantly executes two steps. The first step is a data reduction at each site using torque clustering, and the second step involves performing global clustering with weighted torque clustering. We compare DTC against six state-of-the-art distributed clustering algorithms and automatic centralized clustering techniques on ten large-scale or medium-scale datasets. The results show that the average rank of DTC is at least three times better than those of the other algorithms across all the datasets. Additionally, DTC can accurately predict the ground-truth number of clusters in nine out of ten datasets, further demonstrating its competitive performance and practical potential.

Hierarchical clustering is able to provide partitions of different granularity levels. However, most existing hierarchical clustering techniques perform clustering in the original feature space of the data, which may suffer from overlap, sparseness, or other undesirable characteristics, resulting in noncompetitive performance. In the field of deep clustering, learning representations using pseudo labels has recently become a research hotspot. Yet most existing approaches employ coarse-grained pseudo labels, which may contain noise or incorrect labels. Hence, the learned feature space does not produce a competitive model. In this paper, we introduce the idea of fine-grained labels of supervised learning into unsupervised clustering, giving rise to the enhanced adjacency-constrained hierarchical clustering (ECHC) model. The full framework comprises four steps. One, adjacency-constrained hierarchical clustering (CHC) is used to produce relatively pure fine-grained pseudo labels. Two, those fine-grained pseudo labels are used to train a shallow multilayer perceptron to generate good representations. Three, the corresponding representation of each sample in the learned space is used to construct a similarity matrix. Four, CHC is used to generate the final partition based on the similarity matrix. The experimental results show that the proposed ECHC framework not only outperforms 14 shallow clustering methods on eight real-world datasets but also surpasses current state-of-the-art deep clustering models on six real-world datasets. In addition, on five real-world datasets, ECHC achieves comparable results to supervised algorithms.

The performance of agents is highly influenced by multiple factors, including ability, decision, and states. Trust modeling is widely used to boost the performance of multiagent teaming (MAT). However, most existing trust models rely on statistical methods or preset parameters to assess the trust value in the MAT scenario. In this article, an adaptive trust model is proposed to evaluate comprehensive trust values based on multiple pieces of evidence from variant sources. The proposed trust model leverages information fusion and RL to properly fuse multiple pieces of evidence to generate trust value for every agent in MAT. The trust value is then used in an interaction protocol with MAT to increase the efficiency of cooperation. To verify the performance of the proposed trust model, a ball-collection experiment is designed for MAT to work cooperatively in simulation environments. Two different scenario settings are used to demonstrate the adaptability and robustness of the proposed trust model. The results are further compared with human-designed fusion methods. The comparison shows that the proposed trust model has a better representation of agent performance, namely convergence speed, than human-designed methods in different scenario settings.

The explainability of reinforcement learning (RL) models has received vast amount of interest as its applications have widened. Most existing explainable RL models focus on improving the explainability of an agent's observations instead of the relationships between agent states and actions. This study presents a fuzzy centered explainable network (FCEN) for RL tasks to interpret the relationships between agent states and actions. The proposed FCEN leverages the interpretability of fuzzy neural networks to establish if-then rules and a generative model to visualize learned knowledge. Precisely, the FCEN includes if-then rules that formulate state-action mappings with human-understandable logic, such as the form 'IF Input is A THEN Output is B.' In addition, these rules connect with a generative model that concretizes the states into human-understandable patterns (figures). Our experimental results obtained on 4 Atari games show that the proposed FCEN can achieve a high level of performance in RL tasks and enormously boost the explainability of RL agents both globally and locally. In other words, the FCEN maintains a high-level explanation for the agent decision logic and the possibility of low-level analysis for each given observation sample. The explainability boost does not undermine reward learning performance, humans can even enhance the agent's performance with the provided explainability.

Multisource data fusion analysis, particularly in decision-level fusion strategies, is emerging for application in real-life scenarios. The Dempster-Shafer evidence theory (DSET) is a prevalent approach that has significant importance in managing the fusion tasks. However, existing fusion approaches have limitations in dealing with redundant information and computational complexity associated with the fusion procedure. Though conflict management has been thoroughly studied, other limitations have not been well addressed. In this article, we propose a novel approach for evidential multisource data fusion based on game-theoretic analysis. The introduction of the Shapley function considers the interaction effect of focal elements, mitigating the negative influence of redundant evidence. Additionally, the computational complexity of the fusion procedure is reduced to the same level as the approximate Bayesian update model. We provide a numerical example with conflicting and redundant evidence to show that the proposed approach outperforms current advanced weighted average-based fusion methods. Moreover, a simulation experiment demonstrates the practicality and effectiveness of the proposed approach in identifying driver fatigue states based on electroencephalography (EEG) signals.

Brain-computer interfaces (BCI) help users to interact with machines via brain activity and without the use of muscles. Among the many components of BCI frameworks, sensor technology helps to make the systems highly efficient and robust. As a source of input for BCI systems, sensors also provide high-quality signals that can support downstream tasks, such as noise removal and feature extraction. This book chapter explores the fundamental concepts and configurations of the non-invasive sensor technology that are commonly used in BCI systems.

This paper proposes a novel label ranker network to learn the relationship between labels to solve ranking and classification problems. The Preference Neural Network (PNN) uses spearman correlation gradient ascent and two new activation functions, positive smooth staircase (PSS), and smooth staircase (SS) that accelerate the ranking by creating almost deterministic preference values. PNN is proposed in two forms, fully connected simple Three layers and Preference Net (PN), where the latter is the deep ranking form of PNN to learning feature selection using ranking to solve images classification problem. PN uses a new type of ranker kernel to generate a feature map. PNN outperforms five previously proposed methods for label ranking, obtaining state-of-the-art results on label ranking, and PN achieves promising results on CFAR-100 with high computational efficiency.

Electroencephalograms (EEGs) are brain dynamics measured outside of the brain, which have been widely utilized in non-invasive brain-computer interface applications. Recently, various neural network approaches have been proposed to improve the accuracy of EEG signal recognition. However, these approaches severely rely on manually designed network structures for different tasks which normally are not sharing the same empirical design cross-task-wise. In this paper, we propose a cross-task neural architecture search (CTNAS-EEG) framework for EEG signal recognition, which can automatically design the network structure across tasks and improve the recognition accuracy of EEG signals. Specifically, a compatible search space for cross-task searching and an efficient constrained searching method is proposed to overcome challenges brought by EEG signals. By unifying structure search on different EEG tasks, this work is the first to explore and analyze the searched structure difference in cross-task-wise. Moreover, by introducing architecture search, this work is the first to analyze model performance by customizing model structure for each human subject. Detailed experimental results suggest that the proposed CTNAS-EEG could reach state-of-the-art performance on different EEG tasks, such as Motor Imagery (MI) and Emotion recognition. Extensive experiments and detailed analysis are provided as a good reference for follow-up researchers.

Multisource information fusion is a comprehensive and interdisciplinary subject. Dempster-Shafer (D-S) evidence theory copes with uncertain information effectively. Pattern classification is the core research content of pattern recognition, and multisource information fusion based on D-S evidence theory can be effectively applied to pattern classification problems. However, in D-S evidence theory, highly-conflicting evidence may cause counterintuitive fusion results. Belief divergence theory is one of the theories that are proposed to address problems of highly-conflicting evidence. Although belief divergence can deal with conflict between evidence, none of the existing belief divergence methods has considered how to effectively measure the discrepancy between two pieces of evidence with time evolutionary. In this study, a novel fractal belief Rényi (FBR) divergence is proposed to handle this problem. We assume that it is the first divergence that extends the concept of fractal to R/'enyi divergence. The advantage is measuring the discrepancy between two pieces of evidence with time evolution, which satisfies several properties and is flexible and practical in various circumstances. Furthermore, a novel algorithm for multisource information fusion based on FBR divergence, namely FBReD-based weighted multisource information fusion, is developed. Ultimately, the proposed multisource information fusion algorithm is applied to a series of experiments for pattern classification based on real datasets, where our proposed algorithm achieved superior performance.

The accurate segmentation of medical images plays a crucial role in diagnosing and treating diseases. Although many methods now use multimodal joint segmentation, the joint use of segmentation features extracted by multiple models can lead to heterogeneity and uncertainty. Unreasonable fusion methods cannot exploit the advantages of multiple models and still lack good performance in segmentation. Therefore, this study proposes the FTransCNN model, which is composed of a convolutional neural network (CNN) and Transformer and is based on a fuzzy fusion strategy that jointly utilizes the features extracted by a CNN and Transformer through a new fuzzy fusion module. First, the CNN and Transformer act as the backbone network for parallel feature extraction. Second, channel attention is used to promote the global key information of Transformer to improve the feature representation ability, and spatial attention is used to enhance the local details of CNN features and suppress irrelevant regions. Third, the proposed model applies the Hadamard product to model fine-grained interactions between the two branches and uses the Choquet fuzzy integral to suppress heterogeneity and uncertainty in fused features. Fourth, FTransCNN employs fuzzy attention fusion module (FAFM) hierarchical upsampling to effectively capture both low-level spatial features and high-level semantic context. Finally, the new model obtains the final segmentation result by using the deconvolution and results in an improvement in segmentation. The experimental results on Chest X-ray and Kvasir-SEG dataset show that FTransCNN has better performance on segmentation tasks than the-state-of-the-art deep segmentation models.

The multiple-target self-organizing pursuit (SOP) problem has wide applications and has been considered a challenging self-organization game for distributed systems, in which intelligent agents cooperatively pursue multiple dynamic targets with partial observations. This work proposes a framework for decentralized multi-agent systems to improve the implicit coordination capabilities in search and pursuit. We model a self-organizing system as a partially observable Markov game (POMG) featured by large-scale, decentralization, partial observation, and noncommunication. The proposed distributed algorithm–fuzzy self-organizing cooperative coevolution (FSC2) is then leveraged to resolve the three challenges in multi-target SOP: distributed self-organizing search (SOS), distributed task allocation, and distributed single-target pursuit. FSC2 includes a coordinated multi-agent deep reinforcement learning (MARL) method that enables homogeneous agents to learn natural SOS patterns. Additionally, we propose a fuzzy-based distributed task allocation method, which locally decomposes multi-target SOP into several single-target pursuit problems. The cooperative coevolution principle is employed to coordinate distributed pursuers for each single-target pursuit problem. Therefore, the uncertainties of inherent partial observation and distributed decision-making in the POMG can be alleviated. The experimental results demonstrate that by decomposing the SOP task, FSC2 achieves superior performance compared with other implicit coordination policies fully trained by general MARL algorithms. The scalability of FSC2 is proved that up to 2048 FSC2 agents perform efficient multi-target SOP with almost 100% capture rates. Empirical analyses and ablation studies verify the interpretability, rationality, and effectiveness of component algorithms in FSC2.

The translation of brain dynamics into natural language is pivotal for brain-computer interfaces (BCIs). With the swift advancement of large language models, such as ChatGPT, the need to bridge the gap between the brain and languages becomes increasingly pressing. Current methods, however, require eye-tracking fixations or event markers to segment brain dynamics into word-level features, which can restrict the practical application of these systems. To tackle these issues, we introduce a novel framework, DeWave, that integrates discrete encoding sequences into open-vocabulary EEG-to-text translation tasks. DeWave uses a quantized variational encoder to derive discrete codex encoding and align it with pre-trained language models. This discrete codex representation brings forth two advantages: 1) it realizes translation on raw waves without marker by introducing text-EEG contrastive alignment training, and 2) it alleviates the interference caused by individual differences in EEG waves through an invariant discrete codex with or without markers. Our model surpasses the previous baseline (40.1 and 31.7) by 3.06% and 6.34%, respectively, achieving 41.35 BLEU-1 and 33.71 Rouge-F on the ZuCo Dataset. This work is the first to facilitate the translation of entire EEG signal periods without word-level order markers (e.g., eye fixations), scoring 20.5 BLEU-1 and 29.5 Rouge-1 on the ZuCo Dataset.

Machine learning methods are extensively used for processing and analysing speech signals by virtue of their performance gains over multiple domains. Deep learning and ensemble learning are the two most commonly used techniques, which results in benchmark performance across different downstream tasks. Ensemble deep learning is a recent development which combines these two techniques to result in a robust architecture having substantial performance gains, as well as better generalization performance over the individual techniques. In this paper, we extensively review the use of ensemble deep learning methods for different speech signal related tasks, ranging from general objectives such as automatic speech recognition and voice activity detection, to more specific areas such as biomedical applications involving the detection of pathological speech or music genre detection. We provide a discussion on the use of different ensemble strategies such as bagging, boosting and stacking in the context of speech signals, and identify the various salient features and advantages from a broader perspective when coupled with deep learning architectures. The main objective of this study is to comprehensively evaluate existing works in the area of ensemble deep learning, and highlight the future directions that may be explored to further develop it as a tool for several speech related tasks. To the best of our knowledge, this is the first review study which primarily focuses on ensemble deep learning for speech applications. This study aims to serve as a valuable resource for researchers in academia and in industry working with speech signals, supporting advanced novel applications of ensemble deep learning models towards solving challenges in existing speech processing systems.

Over the years, Machine Learning models have been successfully employed on neuroimaging data for accurately predicting brain age. Deviations from the healthy brain aging pattern are associated with the accelerated brain aging and brain abnormalities. Hence, efficient and accurate diagnosis techniques are required to elicit accurate brain age estimations. Several contributions have been reported in the past for this purpose, resorting to different data-driven modeling methods. Recently, deep neural networks (also referred to as deep learning) have become prevalent in manifold neuroimaging studies, including brain age estimation. In this review, we offer a comprehensive analysis of the literature related to the adoption of deep learning for brain age estimation with neuroimaging data. We detail and analyze different deep learning architectures used for this application, pausing at research works published to date quantitatively exploring their application. We also examine different brain age estimation frameworks, comparatively exposing their advantages and weaknesses. Finally, the review concludes with an outlook towards future directions that should be followed by prospective studies. The ultimate goal of this paper is to establish a common and informed reference for newcomers and experienced researchers willing to approach brain age estimation by using deep learning models.

Vision-based concrete crack detection has recently attracted significant attention from many researchers. Although promising results have been obtained, especially for deep learning (DL) approaches, it is difficult to maintain the robustness of implemented models when tested on completely new data. A possible reason for this is that the extracted feature from the trained set might not fully characterize the crack in the test set. We propose an interdisciplinary approach to improve the effectiveness of vision-based crack detection by modeling crack propagation using fracture mechanics, simulation, and machine learning (ML). Mathematical models of concrete cracks are obtained using ML on the simulation results. Experiments are conducted on various reputable crack image datasets, emphasizing the correlation between simulated and real-world cracks. The importance of propagation models is verified in a classification task, reporting a significant accuracy enhancement on results of some state-of-the-art detection and segmentation models, i.e., 1.27% on average on participating models, and 5.47% on U-Net. This novel approach is expected to have valuable points for a research area where the data quantity and quality still need to be improved.

Alzheimer's disease (AD) is the most prevalent neurodegenerative disorder and the most common cause of Dementia. Neuroimaging analyses, such as T1 weighted magnetic resonance imaging, positron emission tomography, and the deep learning (DL) approaches have attracted researchers for automated AD diagnosis in the early stages. Therefore, a review is required to understand DL algorithms to develop more efficient AD diagnosis methods. This article discusses a detailed review of automated early AD diagnosis using DL methods published from 2009 to 2022. The novelties of this article include: 1) introducing popular imaging modalities; 2) discussing early biomarkers for AD diagnosis using neuroimaging scans; 3) reviewing the popular online available data sets widely used; 4) systematically describing the various DL algorithms for accurate and early assessment of AD; 5) discussion on advantages and limitations of the DL-based model for AD diagnosis; and 6) provides an outlook toward future trends derived from our critical assessment.

Distributed fuzzy neural networks (DFNNs) have attracted increasing attention recently due to their learning abilities in handling data uncertainties in distributed scenarios. However, it is challenging for DFNNs to handle cases in which the local data are non-independent and identically distributed (non-IID). In this paper, we propose a federated fuzzy neural network (FedFNN) with evolutionary rule learning (ERL) to cope with non-IID issues as well as data uncertainties. The FedFNN maintains a global set of rules in a server and a personalized subset of these rules for each local client. ERL is inspired by the theory of biological evolution; it encourages rule variations while activating superior rules and deactivating inferior rules for local clients with non-IID data. Specifically, ERL consists of two stages in an iterative procedure: a rule cooperation stage that updates global rules by aggregating local rules based on their activation statuses and a rule evolution stage that evolves the global rules and updates the activation statuses of the local rules. This procedure improves both the generalization and personalization of the FedFNN for dealing with non-IID issues and data uncertainties. Extensive experiments conducted on a range of datasets demonstrate the superiority of the FedFNN over state-of-the-art methods. Our code is available online11https://github.com/leijiezhang/FedFNN

Fuzzy membership is an effective approach used in twin support vector machines (SVMs) to reduce the effect of noise and outliers in classification problems. Fuzzy twin SVMs (TWSVMs) assign membership weights to reduce the effect of outliers, however, it ignores the positioning of the input data samples and hence fails to distinguish between support vectors and noise. To overcome this issue, intuitionistic fuzzy TWSVM combined the concept of intuitionistic fuzzy number with TWSVMs to reduce the effect of outliers and distinguish support vectors from noise. Despite these benefits, TWSVMs and intuitionistic fuzzy TWSVMs still suffer from some drawbacks as: 1) the local neighborhood information is ignored among the data points and 2) they solve quadratic programming problems (QPPs), which is computationally inefficient. To overcome these issues, we propose a novel intuitionistic fuzzy weighted least squares TWSVMs for classification problems. The proposed approach uses local neighborhood information among the data points and also uses both membership and nonmembership weights to reduce the effect of noise and outliers. The proposed approach solves a system of linear equations instead of solving the QPPs which makes the model more efficient. We evaluated the proposed intuitionistic fuzzy weighted least squares TWSVMs on several benchmark datasets to show the efficiency of the proposed model. Statistical analysis is done to quantify the results statistically. As an application, we used the proposed model for the diagnosis of Schizophrenia disease.

Organised attacks on a computer system to test existing defences, i.e., penetration testing, have been used extensively to evaluate network security. However, penetration testing is a time-consuming process. Additionally, establishing a strategy that resembles a real cyber-attack typically requires in-depth knowledge of the cybersecurity domain. This paper presents a novel architecture, named deep cascaded reinforcement learning agents, or CRLA, that addresses large discrete action spaces in an autonomous penetration testing simulator, where the number of actions exponentially increases with the complexity of the designed cybersecurity network. Employing an algebraic action decomposition strategy, CRLA is shown to find the optimal attack policy in scenarios with large action spaces faster and more stably than a conventional deep Q-learning agent, which is commonly used as a method for applying artificial intelligence to autonomous penetration testing.

With the development of intelligent technology, autonomous agents are no longer just simple tools; they have gradually become our partners. This paper presents a trust-based human-autonomous teaming (HAT) framework to realize tactical coordination between human and autonomous agents. The proposed trust-based HAT framework consists of human and autonomous trust models, which leverage a fusion mechanism to fuse multiple performance metrics to generate trust values in real-time. To obtain adaptive trust models for a particular task, a reinforcement learning algorithm is used to learn the fusion weights of each performance metric from human and autonomous agents. The adaptive trust models enable the proposed trust-based HAT framework to coordinate actions or decisions of human and autonomous agents based on their trust values. We used a ball-collection task to demonstrate the coordination ability of the proposed framework. Our experimental results show that the proposed framework can improve work efficiency.

Community detection in multiplex networks is an emerging research topic in the field of network science. Existing methods usually ignore the similarities among component layers of a multiplex network when detecting its community structures, which decreases the detection efficiency. In this paper, we decompose the community detection in multiplex networks into two problems and propose a novel algorithm which can detect both the specific community partition for each component layer (layer-level community structure) and the composite community structure shared by all layers. Firstly, by specifying the modularity optimization on a network layer as an optimization task, we model the layer-level community detection as a multi-task optimization problem and employ an evolutionary multi-task optimization algorithm to solve it. In this way, the topology correlations among different layers can be utilized to facilitate the community detection. Secondly, we propose an evolutionary clustering ensemble method to find the composite community structure based on the layer-level community partitions and the multiplex network. The proposed method is tested on both synthetic and real-world benchmark networks and compared with classical and state-of-the-art algorithms. Experimental results show that the proposed algorithm has superior community detection performances on multiplex networks.

Failure to notice salient objects even looking directly at them happens when individuals' attention is preoccupied, known as inattentional blindness (IB). As a form of attentional loss, IB occurrence might cause severe outcomes due to limited cognitive resources. Varied methods have been explored to reduce the IB effect and avoid neglect of critical information. Attenuating attentional loss via aided guidance with different sensory modalities intervention could be a possible way to address this issue. This study investigates how different sensory modalities affect the cognitive performance and IB effect from behaviour and neural changes in the human brain and how could we apply this in attention training for attentional loss improvement. Two experimental sessions were conducted, with a multisensory oddball task designed in virtual reality (VR) as the main task to attract individuals' attention. In session 1, participants responded to the main task without being informed of the unexpected task-irrelevant patterns in the background, while in session 2, they were informed of the unexpected patterns but still attended to the main task. Thus, participants were divided into IB (unaware of the pattern) and Aware (aware of the pattern) groups based on their awareness of patterns in the first session. Our results revealed that this VR-based design successfully induced the IB occurrence, with four out of nine participants reporting being unaware of the unexpected patterns. Further, the multisensory oddball task showed better performance in cross-modal stimuli (visual-auditory, VA) with higher accuracy and shorter reaction time than in uni-modal (A or V) conditions. Interestingly, in session 1, the IB group showed better performance than the Aware group, indicating that the IB group was not distracted during the task since they were unaware of the patterns. These findings supported our aims to explore the impact of different sensory modalities on cognitive performance and prov

This paper investigates the application of using error-related potential (ErrP) based brain-computer interface (BCI) paradigm to control robot movements with implicit commands. ErrP is a neural signal that is automatically evoked when the machine’s behavior deviates from the observer’s expectations. By continuously monitoring the presence of ErrP, the system infers the observer’s reaction toward robot movements and automatically translates them into control commands, allowing the implicit control of robot movements without interfering the observer’s other tasks. However, ErrP-based BCI has a major limitation: the ErrP is evoked after the robot has committed an error, which might be costly or dangerous in contexts such as assembly line or autonomous driving. To address these limitations, we propose a novel robotic design for ErrP-based BCI that allows humans to continuously evaluate the robot’s intentions and intervene earlier, if necessary before the robot commits an error. We evaluate the proposed robotic design and BCI system via an experiment where a ground robot performs a binary target-reaching task. The high classification accuracy (77.57%) demonstrated that the proposed ErrP-based BCI is feasible for human-robot intention communication before the robot commits an error and has the potential to broaden the range of applications for ErrP-based BCIs.

Bio-signals pattern recognition systems can be impacted by several factors with a potential to limit their associated performance and clinical translation. Among these factors, selecting the optimum feature extraction method, that can effectively exploit the interaction between the temporal and spatial information, is the most prominent. Despite the potential of deep learning (DL) models for extracting temporal, spatial, or temporal-spatial information, they are typically restricted by their need for a large amount of training data. The deep wavelet scattering transform (WST) is a relatively recent advancement within the DL literature to replace expensive convolution neural networks models with computationally less demanding methods. However, while some studies have used WST to extract features from biological signals, it has not been investigated before for electromyogram (EMG) and electroencephalogram (EEG) signals feature extraction. To investigate the hypothesis of the usefulness of WST for processing EMG and EEG signals, this study used a tremor dataset collected by the authors from people with tremor disorders. Specifically, the proposed work achieved three goals: (a) study the performance of extracting features from low-density EMG signals (8 channels), using the WST approach, (b) study the effect of extracting the features from high-density EEG signals (33 channels), using WST and study its robustness against changing the spatial and temporal aspects of classification accuracy, and (c) classify tremor severity using the WST method and compare the results with other well-known feature extraction approaches. The classification error rates were significantly reduced (maximum of nearly 12%) compared with other feature sets.

Image caption aims to generate a language description of a given image. The problem can be solved by learning semantic information of visual objects and generating descriptions based on extracted embedding. However, the spatial relationship between visual objects and their static position is not fully explored by existing methods. In this work, we propose a Position-Aware Transformer (PAT) model that extracts both regional and static global visual features and unify both the regional and global by incorporating spatial information aligned to each visual feature. To make a better representation of spatial information and correlation between extracted visual features, we propose and compare three subtle approaches to explore position embedding with spatial relation information explicitly. Moreover, we jointly consider the static global and regional embedding for spatial modeling. Experimental results illustrate that our proposed model achieves competitive performance on the COCO image captioning dataset, where the PAT model could respectively reach 38.7, 28.6, and 58.6 on BLEU-4, METEOR, and ROUGE-L respectively. Extensive experiments suggest that the proposed PAT model could also reach competitive performance on related visual-language tasks including visual question answering (VQA) and multi-modal retrieval. Detailed ablation studies are conducted to report how each part would contribute to the final performance, which could be a good reference for follow-up spatial information representation works.

Most current research has focused on non-tonal languages such as English. However, more than 60world’s population speaks tonal languages. Mandarin is the most spoken tonal languages in the world. Interestingly, the use of tone in tonal languages may represent different meanings of words and reflect feelings, which is very different from non-tonal languages. The objective of this study is to determine whether a spoken Mandarin sentence with or without tone can be distinguished by analyzing electroencephalographic signals (EEG). We first constructed a new Brain Research Center Speech (BRCSpeech) database to recognize Mandarin. The EEG data of 14 participants were recorded, while they articulated pre-selected sentences. To our knowledge, this is the first study to apply the method of asymmetric feature extraction method for speech recognition using EEG signals. This study shows that the feature extraction method of Rational Asymmetry (RASM) can achieve the best accuracy in the classification of cross-subjects. In addition, our proposed Binomial Variable Algorithm methodology can achieve 98.82% accuracy in cross-subject classification. Furthermore, we demonstrate that the use of eight channels ((F7, F8), (C5, C6), (P5, P6), and (O1, O2)) can achieve an accurate of 94.44%. This study explores the neuro-physiological correlation of Mandarin pronunciation, which can help develop a tonal language synthesis system based on BCI in the future.

Twin support vector machines (TSVMs) have been successfully employed for binary classification problems. With the advent of machine learning algorithms, data have proliferated and there is a need to handle or process large-scale data. TSVMs are not successful in handling large-scale data due to the following: 1) the optimization problem solved in the TSVM needs to calculate large matrix inverses, which makes it an ineffective choice for large-scale problems; 2) the empirical risk minimization principle is employed in the TSVM and, hence, may suffer due to overfitting; and 3) the Wolfe dual of TSVM formulation involves positive-semidefinite matrices, and hence, singularity issues need to be resolved manually. Keeping in view the aforementioned shortcomings, in this article, we propose a novel large-scale fuzzy least squares TSVM for class imbalance learning (LS-FLSTSVM-CIL). We formulate the LS-FLSTSVM-CIL such that the proposed optimization problem ensures that: 1) no matrix inversion is involved in the proposed LS-FLSTSVM-CIL formulation, which makes it an efficient choice for large-scale problems; 2) the structural risk minimization principle is implemented, which avoids the issues of overfitting and results in better performance; and 3) the Wolfe dual formulation of the proposed LS-FLSTSVM-CIL model involves positive-definite matrices. In addition, to resolve the issues of class imbalance, we assign fuzzy weights in the proposed LS-FLSTSVM-CIL to avoid bias in dominating the samples of class imbalance problems. To make it more feasible for large-scale problems, we use an iterative procedure known as the sequential minimization principle to solve the objective function of the proposed LS-FLSTSVM-CIL model. From the experimental results, one can see that the proposed LS-FLSTSVM-CIL demonstrates superior performance in comparison to baseline classifiers. To demonstrate the feasibility of the proposed LS-FLSTSVM-CIL on large-scale classification problems, we evaluate

Mental task classification (MTC), based on the electroencephalography (EEG) signals is a demanding brain–computer interface (BCI). It is independent of all types of muscular activity. MTC-based BCI systems are capable to identify cognitive activity of human. The success of BCI system depends upon the efficient feature representation from raw EEG signals for classification of mental activities. This paper mainly presents on a novel feature representation (formation of most informative features) of the EEG signal for the both, binary as well as multi MTC, using a combination of some statistical, uncertainty and memory- based coefficient. In this work, the feature formation is carried out in the two stages. In the first stage, the signal is split into different oscillatory functions with the help of three well-known empirical mode decomposition (EMD) algorithms, and a new set of eight parameters (features) are calculated from the oscillatory function in the second stage of feature vector construction. Support vector machine (SVM) is used to classify the feature vectors obtained corresponding to the different mental tasks. This study consists the problem formulation of two variants of MTC; two-class and multi-class MTC. The suggested scheme outperforms the existing work for the both types of mental tasks classification.

Recently, distributed semi-supervised learning (DSSL) algorithms have shown their effectiveness in leveraging unlabeled samples over distributed networks with multiple interconnected agents. However, existing DSSL algorithms cannot cope with data uncertainties and may suffer from high computation and communication overhead problems. Hence, we propose a distributed semi-supervised fuzzy regression model, called DSFR to tackle these issues with a two-pronged strategy - first, a structure learning with a distributed fuzzy C-means method (DFCM) that identifies the parameters in the antecedent component of fuzzy if-then rules; and, second, a parameter learning with distributed interpolation consistency regularization (DICR) to obtain the parameters in the consequent component. Since DFCM is both distributed and unsupervised, it can thus extract feature representation from both labeled and unlabeled samples among multiple agents. Meanwhile, DICR expands sample space with interpolated unlabeled instances in a distributed scheme and forces decision boundaries to lie in sparse data areas, thus increasing the models robustness. Both DFCM and DICR are implemented following the alternating direction method of multipliers method. Notably, none of the procedures involve backpropagation, so the model converges very quickly. Further, with the benefit of DFCM and DICR, DSFR is highly scalable to large datasets. Experiments on both artificial and real-world datasets show that this approach yields much lower loss values than the current state-of-the-art DSSL algorithms at a fraction of the computation cost. Our code is available online\footnote{\url{https://github.com/leijiezhang/DSFR}}.

In recent years, instance segmentation has become a key research area in computer vision. This technology has been applied in varied applications such as robotics, healthcare and intelligent driving. Instance segmentation technology not only detects the location of the object but also marks edges for each single instance, which can solve both object detection and semantic segmentation concurrently. Our survey will give a detail introduction to the instance segmentation technology based on deep learning, reinforcement learning and transformers. Further, we will discuss about its development in this field along with the most common datasets used. We will also focus on different challenges and future development scope for instance segmentation. This technology will provide a strong reference for future researchers in our survey paper.

In this work we develop moderate deviation functions to measure similarity and dissimilarity among a set of given interval-valued data to construct interval-valued aggregation functions, and we apply these functions in two Motor-Imagery Brain Computer Interface (MI-BCI) systems to classify electroencephalography signals. To do so, we introduce the notion of interval-valued moderate deviation function and, in particular, we study those interval-valued moderate deviation functions which preserve the width of the input intervals. In order to apply them in a MI-BCI system, we first use fuzzy implication operators to measure the uncertainty linked to the output of each classifier in the ensemble of the system, and then we perform the decision making phase using the new interval-valued aggregation functions. We have tested the goodness of our proposal in two MI-BCI frameworks, obtaining better results than those obtained using other numerical aggregation and interval-valued OWA operators, and obtaining competitive results versus some non aggregation-based frameworks.

IN THE above article [1], we detected an error in reporting the 10-fold cross-validation result. The correct 10-fold cross-validation result in Table IV is uploaded in this letter. The corrected result of the cross-validation is consistent with our findings [1] that the EEG data associated with virtual global landmark (VGL) [2], [3] stimuli from the VGL group had an overall improvement in Acc and F1 scores compared to local landmarks from the non-VGL group, where Acc and F1 scores improved averagely 14.89% and 21.77%, respectively. (Table Presented).

Robots for physical Human-Robot Collaboration (pHRC) often need to adapt their admittance and how they operate due to several factors. As the admittance of the system becomes variable throughout the workspace, it is not always straightforward for the operator to predict the robot’s behaviour. Previous work demonstrated that cognitive conflicts can be detected during one-dimensional tasks. This work assesses whether cognitive conflicts can also be detected during 2D tasks in pHRC and a classification problem is formulated. Different robot admittance profiles anticipating the stimulus translated into different levels of cognitive conflict. Several commonly used classification algorithms for EEG signals were evaluated to classify different levels of cognitive conflict. Results demonstrate that cognitive conflict level is lower when the admittance smoothly decreases before unexpected events when compared to conditions in which the admittance abruptly decreases before the stimulus. Among the classification algorithms, Convolutional Neural Network has shown the best results to classify different levels of cognitive conflict. Results suggest the feasibility of adaptive approaches for future pHRC control systems that close the loop on users through EEG signals. The detected human cognitive state can also be used to assess and improve the predictability of Human-Robot teams in various pHRC applications.

Complex evidence theory (CET) is an effective method for uncertainty reasoning in knowledge-based systems with good interpretability that has recently attracted much attention. However, approaches to improve the performance of uncertainty reasoning in CET-based expert systems remains an open issue. One key to performance improvement is the adequate management of conflict from multisource information. In this paper, a generalized correlation coefficient, namely, the complex evidential correlation coefficient (CECC), is proposed for the complex mass functions or complex basic belief assignments (CBBAs) in CET. On this basis, a complex conflict coefficient is proposed to measure the conflict between CBBAs; when CBBAs turn into classic BBAs, the complex correlation and conflict coefficients will degrade into traditional coefficients. The complex conflict coefficient satisfies nonnegativity, symmetry, boundedness, extreme consistency, and insensitivity to refinement properties, which are desirable for conflict measurement. Several numerical examples validate through comparisons the superiority of the complex conflict coefficient. In this context, a weighted discounting multisource information fusion algorithm, which is called the CECC-WDMSIF, is designed based on the CECC to improve the performance of CET-based expert systems. By applying the CECC-WDMSIF method to the pattern classification of diverse real-world datasets, it is demonstrated that the proposed CECC-WDMSIF outperforms well-known related approaches with higher classification accuracy and robustness.

Trust is key to achieving successful Human-Robot Interaction (HRI). Besides trust of the human co-worker in the robot, trust of the robot in its human co-worker should also be considered. A computational model of a robot's trust in its human co-worker for physical human-robot collaboration (pHRC) is proposed. The trust model is a function of the human co-worker's performance which can be characterized by factors including safety, robot singularity, smoothness, physical performance and cognitive performance. Experiments with a collaborative robot are conducted to verify the developed trust model.