Zonotope-Based Distributed Set-Membership Fusion Estimation for Artificial Neural Networks Under the Dynamic Event-Triggered Mechanism.

This article is concerned with the distributed set-membership fusion estimation problem for a class of artificial neural networks (ANNs), where the dynamic event-triggered mechanism (ETM) is utilized to schedule the signal transmission from sensors to local estimators to save resource consumption and avoid data congestion. The main purpose of this article is to design a distributed set-membership fusion estimation algorithm that ensures the global estimation error resides in a zonotope at each time instant and, meanwhile, the radius of the zonotope is ultimately bounded. By means of the zonotope properties and the linear matrix inequality (LMI) technique, the zonotope restraining the prediction error is first calculated to improve the prediction accuracy and subsequently, the zonotope enclosing the local estimation error is derived to enhance the estimation performance. By taking into account the side-effect of the order reduction technique (utilized in designing the local estimation algorithm) of the zonotope, a sufficient condition is derived to guarantee the ultimate boundedness of the radius of the zonotope that encompasses the local estimation error. Furthermore, parameters of the local estimators are obtained via solutions to certain bilinear matrix inequalities. Moreover, the zonotope-based distributed fusion estimator is obtained through minimizing certain upper bound of the radius of the zonotope (that contains the global estimation error) according to the matrix-weighted fusion rule. Finally, the effectiveness of the proposed distributed fusion estimation method is illustrated via a numerical example.

Sparse-to-dense coarse-to-fine depth estimation for colonoscopy.

Colonoscopy, as the golden standard for screening colon cancer and diseases, offers considerable benefits to patients. However, it also imposes challenges on diagnosis and potential surgery due to the narrow observation perspective and limited perception dimension. Dense depth estimation can overcome the above limitations and offer doctors straightforward 3D visual feedback. To this end, we propose a novel sparse-to-dense coarse-to-fine depth estimation solution for colonoscopic scenes based on the direct SLAM algorithm. The highlight of our solution is that we utilize the scattered 3D points obtained from SLAM to generate accurate and dense depth in full resolution. This is done by a deep learning (DL)-based depth completion network and a reconstruction system. The depth completion network effectively extracts texture, geometry, and structure features from sparse depth along with RGB data to recover the dense depth map. The reconstruction system further updates the dense depth map using a photometric error-based optimization and a mesh modeling approach to reconstruct a more accurate 3D model of colons with detailed surface texture. We show the effectiveness and accuracy of our depth estimation method on near photo-realistic challenging colon datasets. Experiments demonstrate that the strategy of sparse-to-dense coarse-to-fine can significantly improve the performance of depth estimation and smoothly fuse direct SLAM and DL-based depth estimation into a complete dense reconstruction system.

State Estimation for Discrete Time-Delayed Impulsive Neural Networks Under Communication Constraints: A Delay-Range-Dependent Approach.

In this article, a delay-range-dependent approach is put forward to tackle the state estimation problem for delayed impulsive neural networks. A new type of nonlinear function, which is more general than the normal sigmoid function and functions constrained by the Lipschitz condition, is adopted as the neuron activation function. To effectively alleviate data collisions and save energy, the round-robin protocol is utilized to mitigate the occurrence of unnecessary network congestion in communication channels from sensors to the estimator. With the aid of the Lyapunov stability theory, a state observer is constructed such that the estimation error dynamics are asymptotically stable. The observer existence is ensured by resorting to a set of delay-range-dependent criteria which is dependent on both the impulsive time instant and a coefficient matrix. In addition, the synthesis of the observer is discussed by using linear matrix inequalities. Simulations are provided to illustrate the reasonability of our delay-range-dependent estimation approach.

Predicting Demands of COVID-19 Prevention and Control Materials via Co-Evolutionary Transfer Learning.

The novel coronavirus pneumonia (COVID-19) has created great demands for medical resources. Determining these demands timely and accurately is critically important for the prevention and control of the pandemic. However, even if the infection rate has been estimated, the demands of many medical materials are still difficult to estimate due to their complex relationships with the infection rate and insufficient historical data. To alleviate the difficulties, we propose a co-evolutionary transfer learning (CETL) method for predicting the demands of a set of medical materials, which is important in COVID-19 prevention and control. CETL reuses material demand knowledge not only from other epidemics, such as severe acute respiratory syndrome (SARS) and bird flu but also from natural and manmade disasters. The knowledge or data of these related tasks can also be relatively few and imbalanced. In CETL, each prediction task is implemented by a fuzzy deep contractive autoencoder (CAE), and all prediction networks are cooperatively evolved, simultaneously using intrapopulation evolution to learn task-specific knowledge in each domain and using interpopulation evolution to learn common knowledge shared across the domains. Experimental results show that CETL achieves high prediction accuracies compared to selected state-of-the-art transfer learning and multitask learning models on datasets during two stages of COVID-19 spreading in China.

Partial-Neurons-Based Proportional-Integral Observer Design for Artificial Neural Networks: A Multiple Description Encoding Scheme.

This article is concerned with a new partial-neurons-based proportional-integral observer (PIO) design problem for a class of artificial neural networks (ANNs) subject to bounded disturbances. For the purpose of improving the reliability of the data transmission, the multiple description encoding mechanisms are exploited to encode the measurement data into two identically important descriptions, and the encoded data are then transmitted to the decoders via two individual communication channels susceptible to packet dropouts, where Bernoulli-distributed stochastic variables are utilized to characterize the random occurrence of the packet dropouts. An explicit relationship is discovered that quantifies the influences of the packet dropouts on the decoding accuracy, and a sufficient condition is provided to assess the boundedness of the estimation error dynamics. Furthermore, the desired PIO parameters are calculated by solving two optimization problems based on two metrics (i.e., the smallest ultimate bound and the fastest decay rate) characterizing the estimation performance. Finally, the applicability and advantage of the proposed PIO design strategy are verified by means of an illustrative example.

Kernel Adaptive Filtering Over Complex Networks.

This brief is concerned with the problem of kernel adaptive filtering for a complex network. First, a coupled kernel least mean square (KLMS) algorithm is developed for each node to uncover its nonlinear measurement function by using a series of input-output data. Subsequently, an upper bound is derived for the step-size of the coupled KLMS algorithm to guarantee the mean square convergence. It is shown that the upper bound is dependent on the coupling weights of the complex network. Especially, an optimal step size is obtained to achieve the fastest convergence speed and a suboptimal step size is presented for the purpose of practical implementations. Besides, a coupled kernel recursive least square (KRLS) algorithm is further proposed to improve the filtering performance. Finally, simulations are provided to verify the validity of the theoretical results.

Semi-Supervised Multiview Feature Selection With Adaptive Graph Learning.

As data sources become ever more numerous with increased feature dimensionality, feature selection for multiview data has become an important technique in machine learning. Semi-supervised multiview feature selection (SMFS) focuses on the problem of how to obtain a discriminative feature subset from heterogeneous feature spaces in the case of abundant unlabeled data with little labeled data. Most existing methods suffer from unreliable similarity graph structure across different views since they separate the graph construction from feature selection and use the fixed graphs that are susceptible to noisy features. Furthermore, they directly concatenate multiple feature projections for feature selection, neglecting the contribution diversity among projections. To alleviate these problems, we present an SMFS to simultaneously select informative features and learn a unified graph through the data fusion from aspects of feature projection and similarity graph. Specifically, SMFS adaptively weights different feature projections and flexibly fuses them to form a joint weighted projection, preserving the complementarity and consensus of the original views. Moreover, an implicit graph fusion is devised to dynamically learn a compatible graph across views according to the similarity structure in the learned projection subspace, where the undesirable effects of noisy features are largely alleviated. A convergent method is derived to iteratively optimize SMFS. Experiments on various datasets validate the effectiveness and superiority of SMFS over state-of-the-art methods.

Zonotopic set-membership state estimation for multirate systems with dynamic event-triggered mechanisms.

This paper is concerned with the dynamic event-triggered set-membership state estimation issue for a class of multirate networked systems subject to unknown but bounded disturbance and noise. To economize the limited communication resource when performing the desired networked estimation task, a dynamic event-triggered transmission mechanism is proposed to significantly reduce the frequency of sensor data transmissions. A key issue of the addressed problem is to construct a zonotopic outer approximation set to bound the set of states that are consistent with the disturbed system model and the noisy measured outputs. With the help of a prediction-update-correction method, we design a zonotope that is guaranteed to enclose all possible system states due to the unknown disturbance and noise. Then, the size of such a zonotope is minimized by properly designing an introduced correlation matrix at each iteration. Subsequently, the estimation performance is analyzed under the developed zonotopic estimation method. A sufficient condition is further established to ensure the boundedness of the size of the obtained zonotope. Finally, an illustrative example is provided to demonstrate the effectiveness of the proposed method.

A Differential Evolution Algorithm With Adaptive Niching and K-Means Operation for Data Clustering.

Clustering, as an important part of data mining, is inherently a challenging problem. This article proposes a differential evolution algorithm with adaptive niching and k -means operation (denoted as DE_ANS_AKO) for partitional data clustering. Within the proposed algorithm, an adaptive niching scheme, which can dynamically adjust the size of each niche in the population, is devised and integrated to prevent premature convergence of evolutionary search, thus appropriately searching the space to identify the optimal or near-optimal solution. Furthermore, to improve the search efficiency, an adaptive k -means operation has been designed and employed at the niche level of population. The performance of the proposed algorithm has been evaluated on synthetic as well as real datasets and compared with related methods. The experimental results reveal that the proposed algorithm is able to reliably and efficiently deliver high quality clustering solutions and generally outperforms related methods implemented for comparisons.

Simultaneous State and Unknown Input Estimation for Complex Networks With Redundant Channels Under Dynamic Event-Triggered Mechanisms.

This article addresses the simultaneous state and unknown input estimation problem for a class of discrete time-varying complex networks (CNs) under redundant channels and dynamic event-triggered mechanisms (ETMs). The redundant channels, modeled by an array of mutually independent Bernoulli distributed stochastic variables, are exploited to enhance transmission reliability. For energy-saving purposes, a dynamic event-triggered transmission scheme is enforced to ensure that every sensor node sends its measurement to the corresponding estimator only when a certain condition holds. The primary objective of the investigation carried out is to construct a recursive estimator for both the state and the unknown input such that certain upper bounds on the estimation error covariances are first guaranteed and then minimized at each time instant in the presence of dynamic event-triggered strategies and redundant channels. By solving two series of recursive difference equations, the desired estimator gains are computed. Finally, an illustrative example is presented to show the usefulness of the developed estimator design method.

Finite-Horizon H∞ State Estimation for Stochastic Coupled Networks With Random Inner Couplings Using Round-Robin Protocol.

This article is concerned with the problem of finite-horizon H∞ state estimation for time-varying coupled stochastic networks through the round-robin scheduling protocol. The inner coupling strengths of the considered coupled networks are governed by a random sequence with known expectations and variances. For the sake of mitigating the occurrence probability of the network-induced phenomena, the communication network is equipped with the round-robin protocol that schedules the signal transmissions of the sensors' measurement outputs. By using some dedicated approximation techniques, an uncertain auxiliary system with stochastic parameters is established where the multiplicative noises enter the coefficient matrix of the augmented disturbances. With the established auxiliary system, the desired finite-horizon H∞ state estimator is acquired by solving coupled backward Riccati equations, and the corresponding recursive estimator design algorithm is presented that is suitable for online application. The effectiveness of the proposed estimator design method is validated via a numerical example.

Tridirectional Transfer Learning for Predicting Gastric Cancer Morbidity.

Our previous study has constructed a deep learning model for predicting gastrointestinal infection morbidity based on environmental pollutant indicators in some regions in central China. This article aims to adapt the prediction model for three purposes: 1) predicting the morbidity of a different disease in the same region; 2) predicting the morbidity of the same disease in a different region; and 3) predicting the morbidity of a different disease in a different region. We propose a tridirectional transfer learning approach, which achieves the abovementioned three purposes by: 1) developing a combined univariate regression and multivariate Gaussian model for establishing the relationship between the morbidity of the target disease and that of the source disease together with the high-level pollutant features in the current source region; 2) using mapping-based deep transfer learning to extend the current model to predict the morbidity of the source disease in both source and target regions; and 3) applying the pattern of the combined model in the source region to the extended model to derive a new combined model for predicting the morbidity of the target disease in the target region. We select gastric cancer as the target disease and use the proposed transfer learning approach to predict its morbidity in the source region and three target regions. The results show that, given only a limited number of labeled samples, our approach achieves an average prediction accuracy of over 80% in the source region and up to 78% in the target regions, which can contribute considerably to improving medical preparedness and response.

On the Norm of Dominant Difference for Many-Objective Particle Swarm Optimization.

Recent studies in multiobjective particle swarm optimization (PSO) have the tendency to employ Pareto-based technique, which has a certain effect. However, they will encounter difficulties in their scalability upon many-objective optimization problems (MaOPs) due to the poor discriminability of Pareto optimality, which will affect the selection of leaders, thereby deteriorating the effectiveness of the algorithm. This paper presents a new scheme of discriminating the solutions in objective space. Based on the properties of Pareto optimality, we propose the dominant difference of a solution, which can demonstrate its dominance in every dimension. By investigating the norm of dominant difference among the entire population, the discriminability between the candidates that are difficult to obtain in the objective space is obtained indirectly. By integrating it into PSO, we gained a novel algorithm named many-objective PSO based on the norm of dominant difference (MOPSO/DD) for dealing with MaOPs. Moreover, we design a Lp -norm-based density estimator which makes MOPSO/DD not only have good convergence and diversity but also have lower complexity. Experiments on benchmark problems demonstrate that our proposal is competitive with respect to the state-of-the-art MOPSOs and multiobjective evolutionary algorithms.

Observer-Based Fuzzy Output-Feedback Control for Discrete-Time Strict-Feedback Nonlinear Systems With Stochastic Noises.

This paper focuses on the observer-based output-feedback control (OBOFC) problem for a class of discrete-time strict-feedback nonlinear systems (DTSFNSs) with both multiplicative process noises and additive measurement noises. A state observer is first designed to estimate immeasurable system states, and then a novel observer-based backstepping control framework is proposed for DTSFNSs with known model information. To be specific, virtual control laws and the actual control law are derived using a variable substitution method that gets rid of the repeated accumulation of measurement noises in the recursive process. Furthermore, for technical derivation, the multiplicative noise is successively bounded by state estimation errors and controlled errors. Stability conditions are obtained to guarantee the exponential mean-square boundedness of the closed-loop system. Moreover, the nonlinear modeling uncertainties are taken into account to better reflect engineering practices. In virtue of the universal approximation property of fuzzy-logic systems, a fuzzy observer and the corresponding fuzzy output-feedback controller are simultaneously constructed to derive the stability criteria by using novel weight updated laws. Simulation studies are performed to test the validity of the proposed OBOFC scheme.

Event-Based Adaptive Neural Tracking Control for Discrete-Time Stochastic Nonlinear Systems: A Triggering Threshold Compensation Strategy.

This paper investigates the event-triggered (ET) tracking control problem for a class of discrete-time strict-feedback nonlinear systems subject to both stochastic noises and limited controller-to-actuator communication capacities. The ET mechanism with fixed triggering threshold is designed to decide whether the current control signal should be transmitted to the actuator. A systematic framework is developed to construct a novel adaptive neural controller by directly applying the backstepping procedure to the underlying system. The proposed framework overcomes the noncausality problem, avoids the possible controller-related singularity problem, and gets rid of the neural approximation of the virtual control laws. Under the ET mechanism, the corresponding ET-based actuator is put forward by introducing an ET threshold compensation operator. Such a compensation operator (with an adjustable design parameter) is subtly designed based on a hyperbolic tangent function and a sign function. The threshold compensation error is analytically characterized in terms of a time-varying parameter, and the error bound is shown to be relatively small that is dependent on the adjustable design parameter. Compared with the traditional ET-based actuator without the compensation operator, the proposed ET-based actuator exhibits several distinguished features including: 1) improvement of the tracking accuracy (especially at the triggering instants); 2) further mitigation of the communication load; and 3) enlargement of the allowable range of the ET threshold. These features are illustrated by numerical and practical examples.

Adaptive Neural Event-Triggered Control for Discrete-Time Strict-Feedback Nonlinear Systems.

This paper proposes a novel event-triggered (ET) adaptive neural control scheme for a class of discrete-time nonlinear systems in a strict-feedback form. In the proposed scheme, the ideal control input is derived in a recursive design process, which relies on system states only and is unrelated to virtual control laws. In this case, the high-order neural networks (NNs) are used to approximate the ideal control input (but not the virtual control laws), and then the corresponding adaptive neural controller is developed under the ET mechanism. A modified NN weight updating law, nonperiodically tuned at triggering instants, is designed to guarantee the uniformly ultimate boundedness (UUB) of NN weight estimates for all sampling times. In virtue of the bounded NN weight estimates and a dead-zone operator, the ET condition together with an adaptive ET threshold coefficient is constructed to guarantee the UUB of the closed-loop networked control system through the Lyapunov stability theory, thereby largely easing the network communication load. The proposed ET condition is easy to implement because of the avoidance of: 1) the use of the intermediate ET conditions in the backstepping procedure; 2) the computation of virtual control laws; and 3) the redundant triggering of events when the system states converge to a desired region. The validity of the presented scheme is demonstrated by simulation results.

A Dynamic Event-Triggered Approach to Recursive Filtering for Complex Networks With Switching Topologies Subject to Random Sensor Failures.

This article deals with the recursive filtering issue for a class of nonlinear complex networks (CNs) with switching topologies, random sensor failures and dynamic event-triggered mechanisms. A Markov chain is utilized to characterize the switching behavior of the network topology. The phenomenon of sensor failures occurs in a random way governed by a set of stochastic variables obeying certain probability distributions. In order to save communication cost, a dynamic event-triggered transmission protocol is introduced into the transmission channel from the sensors to the recursive filters. The objective of the addressed problem is to design a set of dynamic event-triggered filters for the underlying CN with a certain guaranteed upper bound (on the filtering error covariance) that is then locally minimized. By employing the induction method, an upper bound is first obtained on the filtering error covariance and subsequently minimized by properly designing the filter parameters. Finally, a simulation example is provided to demonstrate the effectiveness of the proposed filtering scheme.

Mixed H2/H∞ State Estimation for Discrete-Time Switched Complex Networks With Random Coupling Strengths Through Redundant Channels.

This article investigates the mixed H2/H∞ state estimation problem for a class of discrete-time switched complex networks with random coupling strengths through redundant communication channels. A sequence of random variables satisfying certain probability distributions is employed to describe the stochasticity of the coupling strengths. A redundant-channel-based data transmission mechanism is adopted to enhance the reliability of the transmission channel from the sensor to the estimator. The purpose of the addressed problem is to design a state estimator for each node, such that the error dynamics achieves both the stochastic stability (with probability 1) and the prespecified mixed H2/H∞ performance requirement. By using the switched system theory, an extensive stochastic analysis is carried out to derive the sufficient conditions ensuring the stochastic stability as well as the mixed H2/H∞ performance index. The desired state estimator is also parameterized by resorting to the solutions to certain convex optimization problems. A numerical example is provided to illustrate the validity of the proposed estimation scheme.

Generative adversarial network based telecom fraud detection at the receiving bank.

Recently telecom fraud has become a serious problem especially in developing countries such as China. At present, it can be very difficult to coordinate different agencies to prevent fraud completely. In this paper we study how to detect large transfers that are sent from victims deceived by fraudsters at the receiving bank. We propose a new generative adversarial network (GAN) based model to calculate for each large transfer a probability that it is fraudulent, such that the bank can take appropriate measures to prevent potential fraudsters to take the money if the probability exceeds a threshold. The inference model uses a deep denoising autoencoder to effectively learn the complex probabilistic relationship among the input features, and employs adversarial training that establishes a minimax game between a discriminator and a generator to accurately discriminate between positive samples and negative samples in the data distribution. We show that the model outperforms a set of well-known classification methods in experiments, and its applications in two commercial banks have reduced losses of about 10 million RMB in twelve weeks and significantly improved their business reputation.

Airline Passenger Profiling Based on Fuzzy Deep Machine Learning.

Passenger profiling plays a vital part of commercial aviation security, but classical methods become very inefficient in handling the rapidly increasing amounts of electronic records. This paper proposes a deep learning approach to passenger profiling. The center of our approach is a Pythagorean fuzzy deep Boltzmann machine (PFDBM), whose parameters are expressed by Pythagorean fuzzy numbers such that each neuron can learn how a feature affects the production of the correct output from both the positive and negative sides. We propose a hybrid algorithm combining a gradient-based method and an evolutionary algorithm for training the PFDBM. Based on the novel learning model, we develop a deep neural network (DNN) for classifying normal passengers and potential attackers, and further develop an integrated DNN for identifying group attackers whose individual features are insufficient to reveal the abnormality. Experiments on data sets from Air China show that our approach provides much higher learning ability and classification accuracy than existing profilers. It is expected that the fuzzy deep learning approach can be adapted for a variety of complex pattern analysis tasks.