Search

In this paper, we improve the robustness of Machine Learning (ML) classifiers against training-time attacks by linking the risk of training data being tampered with to the redundancy in the ML model's design needed to prevent it. Our defense mechanism is directly applicable to classifiers' training data, without any knowledge of the specific ML model to be hardened. First, we compute the training data proximity to class separation surfaces, identified via a reference linear model. Each data point is associated with a risk index, which is used to partition the training set by an unsupervised technique. Then, we train a learner for each partition and combine the learners' output in an ensemble. Our method treats the protected ML classifier as a black box and is inherently robust to transfer attacks. Experiments show that, for data poisoning rates between 6 and 25 percent of the training set, our method is more robust compared to benchmarks and to a monolithic version of the model trained on the whole training set. Our results make a convincing case for adopting training set partitioning and ensemble generation as a stage of ML models' development and deployment lifecycle. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of secure event-triggered control is to guarantee the stability and security of a general class of industrial cyber-physical systems (CPSs) under limited resource budget and deception attacks. Note that the coupled data and the complicated structure of the CPS make it hard for the traditional control algorithms to fulfill the stability and security requirements. To this end, a neural network learning-based approximation algorithm is first proposed to separate the coupling influence, and then estimate the lumped uncertain nonlinearities. This is done by using the separated data as the input of the neural networks. Second, Nussbaum-type functions are presented to settle the unknown sign of the time-varying attack injection signal. Third, an event triggering mechanism is developed to reduce the communication load. Under the developed secure control laws, all the signals of the resulted closed-loop CPS are bounded. Finally, to validate the effectiveness of the proposed secure event-triggered control method, a benchmark control example for the one-link manipulator actuated by a DC motor is exploited. • Secure event-triggered control is proposed to ensure the security of industrial CPSs. • Neural network learning algorithm is used to estimate the uncertain nonlinearities. • Nussbaum-type functions are presented to handle the unknown sign of the time-varying attack injection signal. • Event triggering mechanism is developed to reduce the communication. [ABSTRACT FROM AUTHOR]

Spectral clustering aims to minimise inter-cluster similarity by constructing graph model, which possesses a significant effect in data of arbitrary shape. Nonetheless, there are still two limitations in the existing algorithms. First, spectral methods perform poorly if the densities of instances vary greatly. Second, they are challenging to handle the complex distribution that different objects of densities are mixed. To address the two limitations, in this paper, a novel spectral clustering algorithm is constructed to accommodate more complex multi-density data. The idea of density stratification is proposed through the probability density of Gaussian mixture model, and the number of layers can be automatically determined by the defined scatter index. Then, a new density ratio is established to effectively reduce the density imbalance of diverse instances based on mutual nearest-neighbour and density stratification, which is more advantageous in the clusters with less clear boundaries. Finally, the optimised adjacency matrix and multi-density spectral clustering algorithm are induced to improve the effect of multi-density data. As per experimental results, the proposed algorithm generally outperforms the popular representative algorithms for both synthetic and benchmark datasets, and works more effectively on multi-density as well as single-density data, which illustrates the superiority of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Recently, 3D deep learning models have been shown to be susceptible to adversarial attacks like their 2D counterparts. Most of the state-of-the-art (SOTA) 3D adversarial attacks perform perturbation to 3D point clouds. To reproduce these attacks in the physical scenario, a generated adversarial 3D point cloud needs to be reconstructed to mesh, which leads to a significant drop in its adversarial effect. In this paper, we propose a strong 3D adversarial attack named Mesh Attack to address this problem by directly performing perturbation on mesh of a 3D object. In order to take advantage of the most effective gradient-based attack, a differentiable sample module that back-propagate the gradient of point cloud to mesh is introduced. To further ensure the adversarial mesh examples without outlier and 3D printable, three mesh losses are adopted. Extensive experiments demonstrate that the proposed scheme outperforms SOTA 3D attacks by a significant margin. We also achieved SOTA performance under various defenses. Our code will be available at: https://github.com/cuge1995/Mesh-Attack. [ABSTRACT FROM AUTHOR]

This issue of DeepFake technique that may cause great threat to privacy, democracy, and national security has attracted the attention of deep learning researchers. DeepFake detection, therefore, has been a very hot issue in deep learning research. The face landmark feature maps are often used by many DeepFake approaches in generating fake faces. It also provides key information to help to detect manipulated face images. In this paper, we propose a detection approach for manipulated face images. To make full use of face landmark information, we propose a facial geometry prior module (FGPM) to extract facial geometry feature maps. Then the facial geometry feature maps are embedded into upsampling feature maps generated by the CNN-LSTM network. The proposed FGPM exploits facial maps and frequency domain correlation to analyze the discriminative characteristics between manipulated and non-manipulated regions by incorporating the CNN-LSTM network. Finally, a decoder is used to learn the mapping from low-resolution feature maps to pixel-wise to predict manipulation localization. Or a softmax classifier is used to predict real or fake face images. By experimenting on several popular datasets, the proposed detection model has demonstrated the capability of localizing manipulation at the pixel level and with a high prediction on real or fake face images. • Propose a manipulated face detection model which predicts fake face in pix-lever. • Facial geometry feature maps are used in predicting fake face. • Propose a CNN-LSTM network to discriminate manipulated face images. [ABSTRACT FROM AUTHOR]

How to derive an investor's preference is vital for investment advisors and online lending platforms for targeted marketing strategies, e.g., market segmentation and financial product recommendation. However, investor preference analysis usually depends on judgments from human investment experts, which are inherently subjective and costly. Intelligent investment advisors (or Robo-advisors), supported by cutting-edge technologies such as machine learning and artificial intelligence, are established to relieve these pending issues. This paper employs an online optimization framework to obtain investors' preferences for further financial product recommendations. This proposed method allows us to update the investor's preference for newly-arriving data sets and tackle the situation where plenty of missing values in investors' records are present. Unlike the black-box-like machine learning approach, our method can provide more managerial implications regarding why one financial product/service is preferred. Real-world data set from an online financial platform is used to compare the existing approaches and shows the stronger and more stable performance of our method when facing different data-missing types and situations with different missing degrees, followed by a recommendation system application. [ABSTRACT FROM AUTHOR]

This paper proposes an approach for the analysis and comparison of stochastic solvers based on the statistical distribution of their variables. The observed variables of the stochastic solvers are the runtime and number of function evaluations required to reach a (sub)-optimal solution. These variables were measured using a target approach. We extended the conventional approach, which usually predicts the average value of variables, by predicting their statistical distributions. If possible, we can predict, not only the average value, but also all the values of the observed variables according to the identified distribution and given probability. We can also predict the probability of reaching (sub)-optimal solutions according to the variable's values. The approach was empirically validated by comparing solvers for discrete and continuous problems. In our experiment, the differences between predicted and measured values of runtime, number of function evaluations and probability are 10% or less. This proves that the proposed approach is useful and can be used for the analysis and comparison of stochastic solvers. Although, the approach has some limitations, it can mitigate the issue of stochastic solvers being unable to provide (sub)-optimal solutions, and can be used to determine the stopping optimization criteria. • An approach to the comparison of stochastic solvers. • The predictive modeling based on optimal and sub optimal solutions. • The analysis of the distribution of runtime and number of function evaluations. • The predictive models based on distribution of solvers' variables are established. • The predictive models provide probability of reaching the optimal solutions. [ABSTRACT FROM AUTHOR]

In online social networks, trust prediction is crucial for applications such as user decision-making and high-impact user recommendation. In recent years, various solutions to the trust prediction problem have been proposed in many academic literatures. However, the existing methods only focus on using more available information to achieve better prediction performance, while ignoring the influence of the authenticity of user information and the user subjectivity. In this paper, we apply the sentiment labels to the trust prediction process for the first time, and propose a deep trust prediction method based on sentiment polarity perception, S-DeepTrust. We first use transfer learning to generate sentiment labels for user review data and obtain a sentiment polarity matrix. Next, we design a rating matrix generation algorithm, which weights the sentiment polarity matrix and the original rating matrix to produce a new rating matrix. Then, the Siamese network architecture is used for the first time in the field of trust prediction to judge the trust relationships based on the extracted user preference vectors, which minimizes the representation of trust user pairs. Finally, the experiment results show that our method achieves higher prediction accuracy and is robust to trust relation sparsity compared with existing methods. [ABSTRACT FROM AUTHOR]

The multivariate spatio-temporal data contains complex spatio-temporal background and channel coupling information. Effective extraction of these features is crucial for data generation. In this paper, a separated multiple attention network is proposed, which can capture the correlation of multiple types of variables in the same space-time, different spaces at the same time, and different times in the same space. Meanwhile, a new multiscale loss processing method based on homoscedasticity uncertainty and the assumption of gaussian loss distribution is proposed to balance the numerical scale of each channel loss in training. The experiment shows that our method has better performance and robustness than the classical machine learning model and higher computational efficiency than the physical model. It can generate multivariate intermittent spatial-temporal fields with a maximum lead time of 3 days and multivariate continuous spatial-temporal fields with a maximum lead time of 7 days. • Novel spatio-temporal data generation model is designed based on separated attention. • New method for balance the learning progress of multivariate data generation. • The lead time for spatio-temporal data generation has been extended to 7 days. [ABSTRACT FROM AUTHOR]

This paper studies a distributed L 2 − L ∞ containment control problem with or without considering nonconvex control input constraints. The main challenge is how to address nonconvex constraints while satisfying the L 2 − L ∞ performance index function. A distributed algorithm is proposed using only the local information, and the containment control problem with nonconvex constraints and L 2 − L ∞ performance can be solved. Based on the properties of the projection operator and convexity, the constrained L 2 − L ∞ containment control problem can be solved under the undirected connected graphs. Finally, the effectiveness of the algorithm is verified by a numerical experiment. [ABSTRACT FROM AUTHOR]

This paper focuses on addressing the bipartite tracking control issue for stochastic nonlinear multiagent systems (MASs) with unmeasured states, unknown external disturbances and actuator nonlinearity. During the design procedure, a category of distributed reduced-order extended state observers (RESOs) are constructed to compensate for the total disturbances and estimate unknown states. Moreover, in order to solve the adverse impacts caused by actuator nonlinearity (saturation and dead-zone), an auxiliary variable and the Nussbaum gain method are adopted. Compared with the existing results, the proposed scheme has a better approximation effect. Based on distributed RESOs, the command-filtered backstepping method, and stochastic Lyapunov stability theory, a bipartite tracking control protocol is designed such that all signals of the closed-loop system are semi-global uniformly ultimately bounded (SGUUB) in probability. Finally, the effectiveness and superiority of the proposed method are verified via simulation examples and comparison results. [ABSTRACT FROM AUTHOR]

Currently, the majority of solutions for human interactive action recognition typically rely on machine vision methods, thus leading to limited application scenarios for accurate recognition. In this paper, we proposed a novel two-level multi-head attentional human interaction action recognition model based on inertial measurement units. Considering that individual actions may contribute differently to interaction actions, the overall model of weighting individual actions to identify interaction actions is divided into individual scenes and interaction scenes. A multi-head attention mechanism is added to the model to obtain action weighting information in both scenes. We combine a bi-directional gated recurrent unit (BiGRU) and an improved convolutional neural network constructs to fully capture spatiotemporal features. The experimental results show that the established model can accurately recognize seven interactive actions with an average recognition accuracy of 98.73%, which verifies the excellent performance of the proposed method. • A new inertial sensor-based action recognition model for human interaction. • A dataset of inertial sensor-based human interaction movements was collected. • Human interaction scenes are divided into individual scenes and interaction scenes. • Multi-head attention mechanism captures local and global information. [ABSTRACT FROM AUTHOR]

• Establish formulas to compute spread and indeterminacy indices of fuzzy judgments. • Develop a novel trapezoidal fuzzy extension of Saaty's consistency. • Propose a new framework to normalize trapezoidal fuzzy utility vectors. • Build a linear program to acquire trapezoidal fuzzy utility vectors from TFRPMs. • Find an analytical solution for optimal fuzzy utility vectors of consistent TFRPMs. Fuzzy extensions of Saaty's consistency play an important role in developing fuzzy analytic hierarchy process based decision-making methods. This paper proposes formulas to calculate ratio-based increasing and decreasing spread indices and to obtain ratio-based indeterminacy indices of modal and support intervals for fuzzy judgments of trapezoidal fuzzy reciprocal preference matrices (TFRPMs). A transitivity equation with parametric fuzzy elements is introduced and combined with the existence of parameter values to define consistency of TFRPMs. To cope with the challenge of identifying the existence, computational formulas are developed to directly obtain values of parameters from the fuzzy judgments in a TFRPM, and an equivalent consistency concept is proposed for TFRPMs. Properties of consistent TFRPMs are presented and a logarithmic deviation based formula is provided to compute consistency indices of TFRPMs. A new framework of normalized vectors with trapezoidal fuzzy elements is introduced. A logarithmic goal programming model is established and transformed into a linear program for acquiring normalized trapezoidal fuzzy utility vectors from TFRPMs. An analytical solution is found for optimized trapezoidal fuzzy utility vectors of consistent TFRPMs. The developed models are validated by three numerical illustrations possessing comparative studies. [ABSTRACT FROM AUTHOR]

Finding k -edge connected components (k -ECCs) is widely used to investigate the major cohesive structures within a graph. Traditional methods utilizing global minimum cut to split the graph iteratively are too slow to deal with graphs of a large size. Recently, approximate algorithms based on a decomposition framework have been proposed to speed up the process. However, these algorithms produce low-quality results in graphs containing vertex pairs that are k -connected but do not belong to a k -ECC. In this paper, we propose a decomposition framework that utilizes a local edge connectivity check method to obtain high-quality results in a shorter period. The proposed method utilizes an improved maximum flow algorithm to detect k -ECCs from local k -cores by merging more vertices in each iteration than existing methods. Moreover, a pruning strategy is applied in the decomposition framework to reduce unnecessary branches in the decomposition tree. Our experimental results on both synthetic and real graphs show that our proposed algorithm is faster than the state-of-the-art algorithms in most datasets and achieves effective results among the approximate algorithms with a 96% accuracy on our experimental datasets. • An efficient local edge connectivity check based decomposition framework to detect k -ECCs with high accuracy of results. • A k -core based maximal merging and a pruning strategy for optimizing the algorithm processing time and the height of decomposition tree. • An improved augmenting path flow method to directly calculate the minimum cut of two vertices in the check procedure. • Experimental results show that the algorithm can detect k -ECCs with outstanding speed while producing high-quality results (96% accuracy). [ABSTRACT FROM AUTHOR]

Outlier detection is a crucial task for identifying unexpected patterns, errors, and behaviors; therefore, maximizing the valuable information obtained from ubiquitous, incomplete, redundant, noisy, and mixed data poses a great challenge. To achieve efficient graph-based outlier detection, we enhance the connectivity between similar objects and weaken the connectivity between heterogeneous objects. The network structure proposed in this paper is called "an i ncomplete l ocal and g lobal n eighborhood i nformation (ILGNI) network." In this network, incomplete mixed data can be exploited considering two aspects; single-attribute local information and multi-attribute global information. Specifically, we initially utilize unsupervised attribute reduction methods to improve data quality. Then, from the perspective of local and global information, we use the level of similarity of objects to design strong-neighborhood and weak-similarity relations to deal with incomplete data. On this basis, the topology reconstruction of a high number of fine-grained neighborhood information networks can be realized. Finally, the outlier scores are calculated considering the stationary distribution of the Markov random walk model over the ILGNI network. The experiments conducted on four real-telecom fraud datasets demonstrate that the proposed algorithm can achieve enhanced outlier detection performance with low time complexity. In addition, the proposed method can effectively mine the information obtained from incomplete data and has high applicability to feature-related and feature-independent datasets. [ABSTRACT FROM AUTHOR]

This paper investigates the issue of fault estimation (FE) in a distributed way for nonlinear time-varying systems via a digital communication channel. In order to relieve the communication limitation, a bit rate allocation scheme is made using data transmitted from sensors to estimators. Next, distributed estimators are designed in terms of the obtained decoded measurement outputs such that an upper bound of estimation error covariance (EEC) is ensured by resorting to the well-known mathematical induction approach. Moreover, the desired gains of constructed estimators are calculated at each iteration by minimizing the upper bound. Subsequently, a sufficient condition is presented to check the uniform boundedness of the designed recursive approaches of the distributed FE. Moreover, the monotonicity with respect to the bit rate is discussed for the fixed estimator gains. An example of Chua's circuits is finally presented to check the applicability of the developed distributed strategy. • A novel bit rate allocation scheme is taken into account for the distributed estimation problem in a recursive way. • The desired estimator gains are acquired by means of minimizing an upper bound • The analysis of uniform boundedness is discussed by means of the classical mathematical induction. • The monotonicity with respect to the bit rate is disclosed. [ABSTRACT FROM AUTHOR]

Current research on machine learning-based intelligent routing focuses on algorithm design and performance optimization. How to deploy it in practice remains a pressing issue. Existing deep learning-based intelligent routing algorithms suffer from a high computational cost, which is hard to be afforded by routers. Considering that the cloud can assist the routers with limited resources to handle complex computation, outsourcing deep learning-based intelligent routing computations to the cloud becomes a feasible solution to support intelligent routing. However, due to the untrustworthiness of the cloud, it is necessary for the routers to verify the truth of outsourced routing results. However, the literature has not yet explored this issue. In this paper, we proposed VeriORouting, a scheme to verify the correctness of outsourced deep learning-based intelligent routing results provided by the cloud. Facing a lazy cloud, VeriORouting allows routers to check the reliability of intelligent routing models by testing model accuracy, and to verify the routing results returned by the cloud without knowing the model by using verification functions generated with multilayer perceptron (MLP) and locality-sensitive hashing (LSH) in advance. We show the robustness of VeriORouting under two attacks raised by the cloud. We evaluate the performance of VeriORouting and compare it with a local intelligent routing scheme. The results show that VeriORouting outperforms the local scheme in terms of computational overhead and storage overhead, especially when the number of routers in the network increases. In terms of communication, VeriORouting reduces communication overhead between routers compared to the local scheme. In addition, we measure the verification performance of VeriORouting under a random attack. VeriORouting achieves a detection success rate of 73% with a false positive rate of 3%, and a detection success rate of 90% with a false positive rate of 25%. • The first scheme to verify deep learning-based intelligent routing outsourced to the cloud. • A mechanism for timely network state information transmission with validation. • A new routing table to reduce redundant executions of verification. • Resistance to two possible attacks from a lazy cloud and advance for practical deployment. [ABSTRACT FROM AUTHOR]

The existing researches on local differential privacy mainly focus on low-dimensional data. These may be faced with high computation cost or low data accuracy when handling high-dimensional data. LoPub published high-dimensional crowd-sourced data, whose precision decreases with the increase of attributes pairs. In this paper, PrivPJ, a method to preserve local differential privacy of published high-dimensional data, is proposed, which can capture a tradeoff between the availability of results and privacy preservation. In PrivPJ, we design mVAE, a multivariate joint distribution algorithm that can alleviate the effect of increasing attributes pairs on accuracy. Then the complete graph is obtained based on the distribution and Markov network, and a local-differential-private junction tree is built based on the graph. PrivPJ can publish a locally private synthetic dataset with the approximate joint distribution. PrivPJ satisfies local differential privacy, and performs well on computation cost, estimation accuracy and misclassification rate of the algorithm on high-dimensional datasets. • The solution achieves the publication of high-dimensional data with local differential privacy. • Keeping the correlations among attributes. • Alleviating the influence of increasing attributes on accuracy and improving the availability of data. • Obtaining joint distribution for discrete and continuous attributes. • The solution captures a tradeoff between results availability and privacy protection. [ABSTRACT FROM AUTHOR]

Gas supply networks play a crucial role in steel enterprises because they provide downstream customers with the gas required for production. In this paper, the real-time scheduling problem of a multi-product gas supply network is first modeled under the framework of reinforcement learning (RL). A safe RL method with prediction and safeguard modules is further developed by utilizing the process knowledge from the gas supply network. The prediction module is designed to predict state changes, and the safeguard module is developed to judge and replace dangerous actions according to current and predicted states. In order to avoid repeated dangerous actions, the safeguard module will provide a negative reward to the agent as feedback whenever a dangerous action is replaced. This active prediction-safeguard strategy is beneficial for reducing trial-and-error costs, speeding up training, and running online. Finally, case studies are implemented on an actual gas supply network to demonstrate the advantages of the proposed method. [ABSTRACT FROM AUTHOR]

Recently, person re-identification (re-ID) has gained much attention for its extensive applications in public security. In order to mine more detailed information, some researches have attempted to extract local representations from important regions through human semantic parsing. However, some valuable visual cues beyond human body, such as backpacks and handbags, which can provide complementary information for recognition, may be misclassified as background noise. In addition, when meeting low-quality captured images or severe occlusions, the semantic regions generated by parsing model are not accurate enough, even resulting in disturbance. In this paper, we present complementary networks for person re-ID, which aims to extract more discriminative and robust local representations with complementary clues. Our model contains two branches and one of which aims to mine potential discriminative information in the background that is useful for identifying a person, as a complement for foreground semantics of human body parts. Another branch is designed to capture the salient regions in global scope so as to make up for the problem of poor discrimination of representations caused by the inaccurate semantic confidence maps of former branch. Experiments show that our model achieves competitive performance in many cases. [ABSTRACT FROM AUTHOR]

Equation discovery, also known as symbolic regression, is a machine learning task of inducing closed-form equations from data and background knowledge. The latter takes various forms. Domain-specific knowledge can constrain the space of candidate equations to those that make sense in the scientific or engineering domain of use. Cross-domain knowledge, on the other hand, imposes general rules for model acceptability, such as parsimony, understandability, or consistency of the equations with the dimensional units of the variables. In this paper, we propose using attribute grammars to ensure the induced equations' dimensional consistency. Attribute grammars are flexible enough to combine cross-domain knowledge on dimensional consistency with domain-specific knowledge expressed as a probabilistic context-free grammar. At the same time, we show that attribute grammars can be efficiently transformed into probabilistic context-free grammars for equation discovery with existing algorithms. Finally, we provide empirical evidence that attribute grammars ensuring dimensional consistency of equations can significantly improve the performance of equation discovery on the standard set of a hundred Feynman benchmarks. • Attribute grammars elegantly express dimensionally consistent expressions. • Efficient combining of dimensional analysis with other background knowledge. • Generate equations by transforming into probabilistic context-free grammar. • Discovered 60% more benchmark equations by accounting for dimensional consistency. [ABSTRACT FROM AUTHOR]

This paper proposes a continuous action iterated dilemma (CAID) in evolutionary game theory with a data-driven compensation network and limited learning ability that considers both players' differences and unknown environment effects. In the traditional dynamic model of CAID, players have identical learning abilities and ignore the influence caused by the environment, which is inconsistent with real society. Therefore, we study the limited learning ability of CAID and the unknown learning mechanism caused by the environment to overcome these problems. Firstly, we propose the dynamic model of limited learning ability for CAID to reveal the law of cooperative evolution in the case when the learning abilities of players are varied. Considering the unknown learning mechanism of players, we adopt the data-driven compensation network to confront the effects of unknown dynamics caused by the environment. In addition, based on the limited learning ability and data-driven compensation network of players, the Lyapunov function is designed to prove the convergence of the CAID, avoiding the high computational complexity caused by the eigenvalues of the Jacobin matrix. In this case, simulations based on two classical dynamic model of evolutionary game theory are carried out to show the effectiveness of our proposed method. [ABSTRACT FROM AUTHOR]

Multi-label learning (MLL) copes with applications such as text classification and image recognition where each instance is associated with multiple labels. Currently, some label-specific feature conversion approaches such as LIFT serve each individual label well, however do not take advantage of label correlation. Some label correlation exploitation approaches such as GLOCAL have independent model prediction and label correlation stages, thus limiting their learning ability. This paper introduces LSTC, a new algorithm that generates label-specific features and exploits third-order label correlation. On the one hand, for each label, we choose the same number of positive and negative representative instances inspired by density peak. These instances are used to generate a new data matrix with label-specific features. On the other hand, we train a paired output prediction network for each label based on these matrices of its own and that of two auxiliary labels. In this way, third-order label correlations are implicitly exploited. Specifically, the two auxiliary labels are the most similar and least similar, respectively, to avoid getting stuck in local optima. Experiments are conducted on seventeen benchmark datasets in comparison with ten popular algorithms. Results on eight measures demonstrate the superiority of LSTC on data from various domains except the text-domain. The source code is available at github.com/fansmale/lstc. • We adopt density peaks under Gaussian kernel to extract label-specific features. • We train the model for each label using the data of two auxiliary labels. • We use the softmax function to obtain numerical predictions from paired outputs. • Our algorithm shows superiority in ranking-based measures of non-text data. [ABSTRACT FROM AUTHOR]

Truth discovery is an efficient technique for tackling data conflict problems in crowd sensing for distributed data collection. As the sensory data to be collected may include sensitive information about users, privacy-preserving truth discovery has attracted significant attention in recent years. Most existing studies apply a centralized architecture based on a cryptographic system, which may be vulnerable to single-point attacks and also has a very high computational cost. In this paper, we propose DPriTD, a decentralized privacy-preserving framework for truth discovery in crowd sensing. The proposed approach leverages the additively homomorphic property of Shamir's Secret Sharing scheme to protect user's privacy. DPriTD provides a strict privacy guarantee for crowd sensing applications. Because each sensitive data point, considered to be a secret, is split into a batch of shares, and the secret cannot be recovered unless a sufficient number of shares are aggregated, DPriTD achieves effective truth discovery while protecting sensitive data from collusion attacks. Furthermore, DPriTD is independent of a centralized server and can perform reliably when not all participants are online in real time. It thus enhances the robustness of a crowd sensing system. Extensive experiments conducted on real-world datasets demonstrate the high performance of our method compared with existing mechanisms. • DPriTD is a decentralized privacy-preserving framework for truth discovery in crowd sensing. • DPriTD leverages the additively homomorphic property of Shamir's Secret Sharing scheme to protect user's privacy. • DPriTD achieves effective truth discovery while protects sensitive data from collusion attacks. [ABSTRACT FROM AUTHOR]

Differential evolution (DE) is one of the widely studied algorithms in evolutionary computation. Recently, many adaptive mechanisms have been proposed for DE including adaptive operator selection and adaptive parameter control. Existing studies consider the two kinds of mechanisms independently. In this paper, we first propose a unified mutation operator with learnable parameters. With different parameter settings, the unified mutation operator degenerates into various classic mutation operators. As a result, by adapting the control parameters of the unified mutation operator, we can realize parameter control and operator selection simultaneously. We then present how to use a neural network to adaptively determine the control parameters. We use natural evolution strategies to train the neural network by modeling the evolutionary process as a Markov decision process. We then embed it into three DEs including classic DE, JADE and LSHADE. Experimental studies show that by embedding the learned unified mutation operator, the performances of these backbone DEs can be improved. Particularly, by embedding the unified mutation operator, LSHADE can perform competitively among state-of-the-art EAs including the winner algorithms in the past CEC competitions. Furthermore, we verify the effectiveness of the unified mutation operator through analyzing the population diversity theoretically. [ABSTRACT FROM AUTHOR]

Recent years have witnessed tremendous progress with regard to few-shot learning methods, especially for classification and segmentation tasks. However, the challenging few-shot image generation has garnered less attention. Training GANs with limited data is a formidable task since discriminator overfitting and memorizing degrade synthesis quality. Existing approaches endeavor to expand the training set with data augmentation techniques. In this paper, we propose ProtoGAN, a GAN that incorporates the metric-learning-based prototype mechanism into adversarial learning. Concretely, we align the prototypes and features of real and generated images in the embedding space, facilitating the fidelity of the generated images. We include a variance term in the objective function in order to promote the synthesis diversity of the generative network. As an intuitive yet effective method, ProtoGAN is easy to implement and is complementary to existing methods. We perform extensive experiments on nineteen datasets spanning a variety of domains and resolutions. Both quantitative and qualitative results demonstrate the effectiveness of our proposed method for high-quality image synthesis under limited data. Our code is available at https://github.com/kobeshegu/ProtoGAN. [ABSTRACT FROM AUTHOR]

This paper explores the fault detection filtering problem of Markov switching memristive neural networks with network-induced constraints in the discrete-time domain. The mode changes of memristive neural networks are described by a piecewise nonhomogeneous Markov process, whose transition probabilities are time-varying and governed by a higher-level nonhomogeneous Markov process. A generalized framework of Markov switching memristive neural networks includes the existing neural networks as special cases. In light of the limited communication bandwidth, the quantized measurement and packet dropouts are considered jointly. A mode-dependent fault detection filter is constructed to generate a residual signal and achieve better performance. From the mode-dependent yet time-varying Lyapunov functional, some less conservative sufficient conditions are devised for Markov switching memristive neural networks to ensure the performance level. Eventually, a simulation example is addressed to verify the feasibility of the attained theoretical analysis. [ABSTRACT FROM AUTHOR]

Wind speed forecasting plays a crucial role in reducing the risk of wind power uncertainty, which is vital for power system planning, scheduling, control, and operation. However, it is challenging to obtain accurate wind speed forecasting results since wind speed series contain complex fluctuations. In this paper, a novel wind speed forecasting model is proposed by using a genetic algorithm (GA) and long short-term memory neural network (LSTM), where GA is used for evolving the architectures and hyper-parameters of the LSTM, called EvLSTM, because there is no clear knowledge to determine these crucial parameters. In the proposed EvLSTM forecasting model, a flexible gene encoding strategy, crossover operation, and mutation operation are proposed to describe the different architectures and hyper-parameters of the LSTM during the evolutionary process. In addition, to overcome the weakness of the single method for wind speed forecasting, the ensemble EvLSTM (EnEvLSTM) is proposed by using the no negative constraint theory ensemble learning strategy, whose weight coefficients are determined by the differential evolution algorithm. The proposed EvLSTM and EnEvLSTM forecasting models are evaluated on two real-world wind farms located in Inner Mongolia, China and Sotavento Galicia, Spain. Experimental results on two forecasting horizons demonstrate the superiority of EvLSTM and EnEvLSTM in terms of three performance indices and three statistical tests. [ABSTRACT FROM AUTHOR]

The three-way decisions strategy has been employed to construct network topology in a single hidden layer feedforward neural network (SFNN). However, this model has a general performance, and does not consider the process costs, since it has fixed threshold parameters. Inspired by the sequential three-way decisions (STWD), this paper proposes STWD with an SFNN (STWD-SFNN) to enhance the performance of networks on structured datasets. STWD-SFNN adopts multi-granularity levels to dynamically learn the number of hidden layer nodes from coarse to fine, and set the sequential threshold parameters. Specifically, at the coarse granular level, STWD-SFNN handles easy-to-classify instances by applying strict threshold conditions, and with the increasing number of hidden layer nodes at the fine granular level, STWD-SFNN focuses more on disposing of the difficult-to-classify instances by applying loose threshold conditions, thereby realizing the classification of instances. Moreover, STWD-SFNN considers and reports the process cost produced from each granular level. The experimental results verify that STWD-SFNN has a more compact network on structured datasets than other SFNN models, and has better generalization performance than the competitive models. All models and datasets can be downloaded from https://github.com/wuc567/Machine-learning/tree/main/STWD-SFNN. • We propose the STWD-SFNN model to optimize the hidden layer nodes of neural network. • STWD-SFNN adopts sequential three-way decisions to determine the network structure. • STWD-SFNN has a more compact network structure than TWD-SFNN and other SFNN models. • STWD-SFNN has a better generalization performance than the competitive models. • We analyze the costs of test and delay of STWD-SFNN. [ABSTRACT FROM AUTHOR]

A hybrid information system (HIS) means an information system (IS) with categorical attributes, real-valued attributes and missing values. It is more difficult to deal with an HIS than an IS with a type of attributes. This paper studies attribute reduction in an HIS by means of fuzzy rough iterative computation model (FRIC-model). To more accurately express the difference between objects in an HIS, a novel distance function on the object set is first constructed. Then, fuzzy symmetric relations on the object set based on this function are established and fuzzy rough approximations are defined. Next, some evaluation functions that represent the classification ability of attribute subsets, such as fuzzy positive regions, dependency functions and attribute importance functions are introduced. After then, FRIC-model is given and an attribute reduction algorithm for hybrid data based on this model is designed. At last, numerical experiments are carried out to estimate the performance of the designed algorithm. The experimental results show that the designed algorithm is more effective than some existing algorithms. [ABSTRACT FROM AUTHOR]

Explainable artificial intelligence (XAI) is gradually becoming a key component of many artificial intelligence systems. However, such pursuit of transparency may bring potential privacy threats to the model confidentially, as the adversary may obtain more critical information about the model. In this paper, we systematically study how model decision explanations impact model extraction attacks, which aim at stealing the functionalities of a black-box model. Based on the threat models we formulated, an XAI-aware model extraction attack (XaMEA), a novel attack framework that exploits spatial knowledge from decision explanations is proposed. XaMEA is designed to be model-agnostic: it achieves considerable extraction fidelity on arbitrary machine learning (ML) models. Moreover, we proved that this attack is inexorable, even if the target model does not proactively provide model explanations. Various empirical results have also verified the effectiveness of XaMEA and disclosed privacy leakages caused by decision explanations. We hope this work would highlight the need for techniques that better trade off the transparency and privacy of ML models. • We propose XaMEA, three XAI-aware model extraction attack architectures. • We further carry out XAI-aware model extraction attacks against non-explanation target models. • We evaluate the attack effectiveness of XaMEA with an exhaustive set of experiments. [ABSTRACT FROM AUTHOR]

Flatness is a crucial indicator of strip quality that presents a challenge in regulation due to the high-speed process and the nonlinear relationship between flatness and process parameters. Conventional methods for controlling flatness are based on the first principles, empirical models, and predesigned rules, which are less adaptable to changing rolling conditions. To address this limitation, this paper proposed an offline reinforcement learning (RL) based data-driven method for flatness control. Based on the data collected from a factory, the offline RL method can learn the process dynamics from data to generate a control policy. Unlike online RL methods, the proposed method does not require a simulator for training, the policy can be potentially safer and more accurate since a simulator involves simplifications that can introduce bias. To obtain a steady performance, the proposed method incorporated ensemble Q-functions into policy evaluation to address uncertainty estimation. To address distributional shifts, based on Q-values from ensemble Q-functions, behavior cloning was added to policy improvement. Simulation and comparison results showed that the proposed method outperformed the state-of-the-art offline RL methods and achieved the best performance in producing strips with lower flatness. [ABSTRACT FROM AUTHOR]

Similarity matrix is of vital importance for graph-based multi-view clustering models, which can depict the nonlinear structure information among samples. However, most existing approaches learn the common similarity graphs for clustering assignment from original data points directly, which shows the unclear structure to result in the inability to accurately extract underlying information in datasets. In this paper, a novel multi-view clustering model called AONGR is proposed, which integrates spectral clustering and nonnegative matrix factorization into a joint framework to reconstruct the similarity graphs. The reconstructed graph not only owns a clear structure but offers the interpretation for cluster assignment. The contribution of each view on clustering can also be obtained during the procedure of optimization. And an effective algorithm is designed to update the variables in AONGR. Experimental results on eight real-world datasets demonstrate the superiority of our AONGR in comparison with the state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

Network dismantling problem aims to find a node subset whose removal from a network results in the fragmentation of the network into subcritical connected components at the minimal overall cost. People have always been more interested in the unweighted case where each node has the same cost, while there are few results for the weighted case when nodes have different costs. It is a much more challenging problem in network science. In this paper, we present a novel strategy for this generalized network dismantling problem by characterizing the relationship between blocks and cut nodes. We firstly construct an auxiliary block-cut tree T from the original network G , where the nodes in the tree T have two types which correspond to either blocks or cut nodes of G , and edges of T only exist between two different type of nodes. Then we transform the problem of partitioning G by removing nodes to the problem of partitioning T by removing edges, both at a minimum overall cost. Finally partitioning T can be solved by a weighted spectral clustering algorithm. When we apply this new strategy on real-world networks, it performs equally or even better than the current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper studies the problem of rescheduling trains with speed management in uncertain disruptions that lead to temporary speed restrictions (TSRs) in high-speed railway systems. The disruption uncertainties are that the coverage and speed limit value of TSRs change randomly over time, and the start and end times of TSRs are unknown in advance. First, a mixed-integer linear programming model is formulated to reduce the total train traveling times and improve the passenger comfortability. The solutions of the model provide simultaneously the optimal train rescheduling strategies and train speed control strategies. Second, a rolling horizon algorithm is applied in consideration of the disruption uncertainties. The rolling horizon algorithm updates and solves the model in every time horizon, according to the disruption information newly detected. Therefore, the train rescheduling orders and speed control orders can be adjusted according to the real-time situation to guarantee effectiveness. The rolling horizon approach is terminated automatically once the disruption ends. Finally, based on the practical data of the Beijing-Tianjin intercity high-speed railway line, numerical experiments are carried out to test the effectiveness and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

Graph convolution neural network (GCN) shows strong performance in non-Euclidean structure data. In recent years, many researchers have applied GCN to Euclidean structure data, such as images and languages, and obtained excellent results, which expanded the application range of GCN. Despite a large number of related methods, few studies deal with information systems from the perspective of graphs. In fact, the information system is also a non-Euclidean data structure, and there is serious information loss when using classical measurement methods to construct its data structure. Thus, the promotion of GCN is hindered in information systems. This paper applies GCN to the classification of information systems. Firstly, an intuitionistic fuzzy relation, called positive-negative relation, is established in the information system, and the physical meaning of this relationship is explained. Secondly, based on the positive-negative relation, a positive-negative relation graph convolution neural network model is constructed called PN-GCN. Finally, the effectiveness and robustness of PN-GCN are validated through experiments. [ABSTRACT FROM AUTHOR]

Consensus reaching process (CRP) is very important for multi-criteria group decision making (MCGDM). Recently, social network driven CRP methods have been deeply investigated. However, these methods rarely considered knowledge levels of DMs. In fact, knowledge levels of DMs can decide the correctness of final decision result which play an important role in CRP. Therefore, this paper proposes knowledge degree to measure knowledge level of DM and develops a new social network driven CRP method for probabilistic linguistic MCGDM problems. Based on the similarity degree and knowledge degree, similarity degree-based social network (SDSN) and trust relationship-based social network (TRSN) are constructed successively. According to the constructed SDSN and TRSN, the proposed CRP method consists of two stages. In the first stage, pair-wise DMs in complementary social network adjust their evaluations. In the second stage, the evaluations of DMs who contribute less to group consensus are modified. To ensure that the group consensus index (GCI) is non-decreasing in the above two stages, the constructed programming models not only improve similarity degrees between DMs-to-adjust and reference DMs but also consider similarity degrees between DMs-to-adjust and other DMs. Finally, an investment selection example and comparison analyses demonstrate the applicability and advantages of the proposed method. [ABSTRACT FROM AUTHOR]

With the vigorous development of big data, the task of large-scale community structure detection is more challenging. In this paper, a large-scale community detection method based on core node and layer-by-layer label propagation is proposed, which is further extended to the detection of overlapping community structures. First, the core nodes whose node degree is greater than the average degree in the graph are found to make effective use of the feature that the core node is the potential community center. This will also avoid the impact of nodes with low node degrees on community structure detection. Then, starting from the core node, label propagation is carried out layer-by-layer according to the node degree and node connection, which effectively improves the accuracy of community detection. The node labels after label propagation are calibrated according to the current attraction of the community to the nodes, which effectively improves the situation of misclassification in the early community detection. Finally, overlapping community detection is carried out based on the non-overlapping community structure. At this time, the overlapping community detection result is more accurate and interpretable. The community detection results on 3 synthetic networks and 12 real datasets show that the proposed algorithm has more advantages than four non-overlapping community detection methods and two overlapping community detection methods. [ABSTRACT FROM AUTHOR]

Graph representation learning is a hot topic in non-Euclidean data in various domains, such as social networks, biological networks, etc. When some data labels are missing, graph autoencoder and graph variational autoencoder can perform outstanding abilities on node clustering or link prediction tasks. However, most existing graph representation learning ignores data's multi-modal features and takes the node features and graph structure features as one view. Besides, most graph autoencoders only reconstruct the node feature matrix or adjacency matrix, which does not fully use the hidden representation information. In this paper, we propose a multi-view representation model based on graph autoencoder, which can employ the global structure topology, latent local topology, and feature relative information. Meanwhile, we add another decoder to reconstruct the node feature matrix as an auxiliary task. In this way, the proposed framework can utilize the learned representation sufficiently. We validate the effectiveness of our framework on four datasets, and the experimental results demonstrate superior performance compared with other advanced frameworks. [ABSTRACT FROM AUTHOR]

Semi-supervised clustering (SSC), a technique integrating semi-supervised learning and clustering analysis, incorporates the given prior information (e.g., class labels and pairwise constraints) into clustering to guide the clustering process and improve the performance. In recent years, a large number of valuable works have emerged, focusing on theoretical research and application in different fields. In this paper, a detailed review of SSC is provided from a new perspective. Firstly, all SSC studies are organized as partition-based SSC , hierarchical-based SSC , density-based SSC , graph-based SSC , neural network-based SSC , Nonnegative Matrix Factorization-based SSC and random subspace technique-based SSC. Thus, the semi-supervised researches can be in-depth discussed in each clustering idea. Secondly, the general overviews are detailed in each category respectively, including the performance, the suitable scenarios and the way to add supervising information. Thirdly, the recent successful applications of SSC are summarized according to different backgrounds such as medical, biological, business, journalism, financial and so on. Based on this, some application caveats and development trends of SSC are particularly given in the end. This comprehensive review and analysis of SSC can provide an overall outline, the scope of research topics, and a relative complete analysis of existing SSC methods for researchers. [ABSTRACT FROM AUTHOR]

Public bicycle sharing systems (PBSS) allow citizens to move between two points within a region using bicycles, being composed of stations where users can either pick up or return bicycles after using them. Depending on the demand for bicycles, stations may either become empty or full of bicycles. This problem is usually solved by means of a vehicle that exchanges bicycles among the stations following some balancing strategy. In this paper, we propose a model for PBSSs using generalized stochastic Petri nets with external control, a Petri net with an external agent that decides how to execute the balancing strategy. The main advantage of our model is that it allows different balancing strategies to be implemented without changing the Petri net structure. In addition, since the model is a Petri net, we can easily implement it by making a Petri net simulator built with the incidence matrix associated with the Petri net model. We also propose two balancing strategies and validate both the simulation algorithm and the balancing strategies by applying them to the performance evaluations of the model of a real system, named Integra UFRJ PBSS. The performance evaluation of the system under the balacing strategies have shown that both strategies considerably reduce the times, in average, that stations remain full or empty throughout the system operation, solving problems related to stations being empty of full; in particular, the first strategy reduced those times in such a way that no more than five percent of the users of each station had trouble returning their bicycles due to stations being full, whereas the second strategy reduced the percentage of lost users due to stations being empty to no more than five percent for each station. [ABSTRACT FROM AUTHOR]

Existing multi-stage trackers treat visual object tracking as a multiple feature extraction and similarity metric process. However, the similarity metric methods used in them are typically based on linear cross-correlation, ignoring the matching of detailed information. Moreover, the feature extraction operators (e.g., RoI align) lead to a sub-optimal feature representation for matching. In this paper, we propose a novel similarity metric method called feature alignment-based multi-level similarity metric learning network to address these issues. Technically, we elaborate a feature alignment module to extract the features, suppressing the useless background information that affects the matching. Subsequently, using the aligned features, we design a learnable multi-level similarity metric learning network to implement the matching for detailed information at the channel and spatial levels, which effectively guides an accurate and discriminative similarity score. By integrating the above components as second stage, a two-stage robust tracking method FAML-RT is presented. Extensive experiments on the challenging benchmarks OTB100, LaSOT and VOT2018 show that FAML-RT achieves a competitive performance against state-of-the-art methods, while running at a high speed of 60 fps. Furthermore, a series of ablation studies demonstrate the effectiveness of the proposed feature alignment-based multi-level metric learning network. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the impact of optimal hyper-parameter configuration in relational topic models. The main goal is to validate the hypothesis that single-objective Bayesian Optimization (BO) can discover a hyper-parameter setting that leads a set of relational topic models to simultaneously ensure good prediction capabilities and significant topics from a qualitative perspective. Our research, as a result of a comparative analysis performed on 7 state-of-the-art models, 5 performance measures and 3 datasets, has highlighted three main findings: (1) the majority of relational topic models are not able to offer a good trade-off between classification capabilities and topic interpretability; (2) single-objective optimization of hyper-parameters, targeted on maximizing the F1-Measure, is able to create topics that are also optimal with respect to the Kullback Leibler divergence measure; (3) the Pareto frontiers across several performance metrics reveals that the most promising trade-off between the performance metrics can be obtained by Constrained Relational Topic Models. [ABSTRACT FROM AUTHOR]

Appendable-block blockchain was established in an effort to standardize a blockchain architecture for heterogeneous Internet of Things devices. In current constructions, the main disadvantage lies in the full trust of gateways, which can launch coordinated and individual eclipse attacks against IoT sensing devices without being caught. This paper introduces Hieraledger, a novel commitment protocol for exposing malicious gateways in appendable-block blockchain constructions. To minimize the likelihood of eclipse attacks, each sensing device selects more than one gateway based on a novel reputation scoring model and the self-valuation of its privacy. To expose malicious gateways, our protocol employs fail-stop signatures thwarting signature forging, cryptographic hash functions, and a novel Merkle appendable tree. For each sensing device, we generate a resource-aware repudiable cryptographic proof of origin evidence. For gateway nodes, we generate a chain of commitment-based non-repudiable cryptographic proofs of delivery evidence. To demonstrate the security and robustness of our scheme, we analyze it under a solid theoretical security model. Based on our findings, our protocol enables sensing devices to seek arbitration and expose malicious gateways to the network. The present study is a further addition to the growing body of research into privacy preserving blockchain solutions for the Internet of Things [ABSTRACT FROM AUTHOR]

A fault estimation and compensation approach is presented for discrete-time systems with event-triggered scheme based on zonotope techniques in this paper. Instead of energy-bounded uncertainties, the dynamic system is supposed to be subjected to unknown but amplitude-bounded disturbance and noise, which is more in line with engineering practice. By introducing dynamic interval variables, a component-wise dynamic event-triggered strategy is utilized to reduce the consumption of communication resources. With considering the amplitude-bounded uncertainties and event-triggered scheme, a joint state-fault estimator is constructed to estimate the system state as well as possible system faults simultaneously, which are utilized to construct the fault-tolerant controller by compensating the effects of system faults. A co-design approach of the estimator, compensator, and event-triggered strategy is proposed to guarantee the stability and l 1 performance of the system, based on which a zonotope-based interval estimation algorithm of system state and possible faults is provided to obtain the tight estimation of system state and faults. Eventually, simulation results on aircraft engine systems indicate that the designed method achieves a satisfactory fault estimation and fault tolerance performance, and also reduce the consumption of communication resources significantly. [ABSTRACT FROM AUTHOR]

Auction, widely used by firms and governments, has facilitated trillions of dollars in the world economy. However, when price is the only factor determining the winner, how to guarantee the trustworthiness of the auction outcomes, and at the same time preserve the privacy of losing bidders is a challenging issue. In this paper, to address this issue, we present a sealed-bid auction protocol with privacy preserving and public verifiability atop blockchain. No bidder is able to repudiate or change any bidding to outbid others and all public information comes with proof that even observers are able to validate the auction. Proofs post along the way to validate biddings and identify the winner are in the form of group elements with communication complexity O (log ⁡ (n m)) , n , m denoting the upper bound bit length of the bidding price and the number of bidders respectively. The logarithmic complexity significantly reduces the communication cost and makes it practical for large-scale auction. Moreover, our protocol can be easily extended to any type of ordinal-price sealed-bid auction such as second-price with privacy preserving. We provide detailed security proofs or analysis of the required security goals. The evaluation of time cost in off-chain operations and gas cost in on-chain functions demonstrate the practicality of our protocol in real-world applications. • Propose a blockchain-based trustworthy sealed-bid auction protocol with low communication cost. • Provide a method for validation of bidding and the trustworthiness of the outcome. • Prove a detailed security proof for the protocol. • Evaluate the time cost of off-chain functions and the gas cost of on-chain functions. [ABSTRACT FROM AUTHOR]

In recent years, researchers have proposed many graph-based multi-view clustering (GMC) algorithms to solve the multi-view clustering (MVC) problem. However, there are still some limitations in the existing GMC algorithm. In these algorithms, a graph is usually constructed to represent the relationship between the samples in a view; however, it cannot represent the relationship very well since it is often preconstructed. In addition, these algorithms ignore the robustness problem of each graph and high-level information between different graphs. Then, in the paper, we propose a novel MVC method, i.e., robust and optimal neighborhood graph learning for MVC (RONGL/MVC). Specifically, we first build an initial graph for each view. However, these initial graphs cannot represent the relationship between the samples in each view well, so we look for the optimal graph with k connected components in the neighborhood of each initial graph, where k is the number of clusters. Then, to improve the robustness of RONGL/MVC, we reconstruct the optimal graph with the self-representation matrix. Furthermore, we stack all the self-representation matrices into a tensor and impose the tensor low-rank constraint, which can maximize consistent features and explore the high-order relationship between optimal graphs. In addition, we provide an optimization strategy by utilizing the Augmented Lagrange Multiplier (ALM) method. Experimental results on several datasets indicate that RONGL/MVC outperforms state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Ambient intelligence (AmI) systems aim to provide users with context-aware assistance services intended to improve the quality of their lives in terms of autonomy, safety, and well-being. Taking the uncertainty and partial observability of these environments into account is of major importance for context recognition and, more specifically, to detect and solve context abnormalities such as those related to the user's behavior or those related to context attribute prediction. In this paper, an ontology-based framework integrating machine learning and probabilistic planning within commonsense reasoning is proposed to recognize the user's context and abnormalities associated with it. The reasoning is performed using event calculus in answer set programming (ECASP); ECASP allows for abductive and temporal reasoning, which results in an eXplainable AI (XAI) approach. A context ontology is proposed to axiomatize the reasoning and introduce the notion of probabilistic fluents into the EC formalism in order to perform probabilistic reasoning. The reasoning incorporates probabilistic planning based on a partially observable Markov decision process (POMDP) to solve knowledge incompleteness. To evaluate the proposed framework, real-life scenarios, based on the Orange4Home and SIMADL public datasets are implemented and discussed. [ABSTRACT FROM AUTHOR]

Uncovering the reward function of optimal controllers is crucial to determine the desired performance that an expert wants to inject to a certain dynamical system. In this paper, a reward inference algorithm of discrete-time expert's controllers is proposed. The approach is inspired by the complementary mechanisms of the striatum, neocortex, and hippocampus for decision making and experience transference. These systems work together to infer the reward function associated to expert's controller using the complementary merits of data-driven and online learning methods. The proposed approach models the neocortex system as two independent learning algorithms given by a Q-learning algorithm and a gradient identification rule. The hippocampus is modelled by a least-squares update rule that extracts the relation from the states and control inputs of the expert's data. The striatum is modelled by an inverse optimal control algorithm which iteratively finds the hidden reward function. Lyapunov stability theory is used to show the stability and convergence of the proposed approach. Simulation studies are given to demonstrate the effectiveness of the proposed complementary learning algorithm. [ABSTRACT FROM AUTHOR]

Distributed estimation for parametric models has drawn attention in modern statistical learning, but few studies focus on semiparametric models. In this paper, we propose two communication-efficient distributed estimators for partially linear additive models with high-dimensional covariates. The commonly used B-spline basis functions are first applied to approximate the nonparametric functions and then we construct a profiled communication-efficient surrogate loss function with Lasso penalty based on one local machine solving the final optimization problem. Further, to reduce the effect of local machines and improve the stability of the algorithm, a profiled gradient-enhanced loss estimator is derived. The resulting two estimators and their theoretical convergence rates for both parametric and nonparametric components are established. The finite-sample performance of the proposed estimators is studied through simulations and an application to appliances energy prediction data set is also presented. [ABSTRACT FROM AUTHOR]