Search

Linear discriminant analysis (LDA) aims to find a low-dimensional space in which data points in the same class are to be close to each other while keeping data points from different classes apart. To improve the robustness of LDA to non-Gaussian distribution data, most existing discriminant analysis methods extend LDA by approximating the underlying manifold of data. However, these methods suffer from the following problems: 1) local affinity or reconstruction coefficients are learned on the basis of the relationships of all data pairs, which would lead to a sharp increase in the amount of computation and 2) they learn the manifold information in the original space, ignoring the interference of the noise and redundant features. Motivated by these challenges, this article represents a novel discriminant analysis model, called fast and adaptive locality discriminant analysis (FALDA), to improve the efficiency and robustness. First, with the anchor-based strategy, a bipartite graph of each class is constructed to characterize the local structure of data. Since the number of anchor points is far less than that of data points, learning of fuzzy membership relationships between data points and anchor points within each class can save training time. Second, a maximum entropy regularization is introduced to control the uniformity of the weights of graphs and avoid the trivial solution. Third, the above relationships are updated adaptively in the process of dimensionality reduction, which can suppress the interference of the noise and redundant features. Fourth, the whitening constraint is imposed on the projection matrix to remove the relevance between features and restrict the total scatter of data in the subspace. Last but not the least, data with complex distribution can be explicitly divided into sub-blocks according to the learned anchor points (or subclass center points). We test our proposed method on synthetic data, benchmark datasets, and imbalanced datasets. Promising experimental results demonstrate the success of this novel model.

Der Beitrag analysiert, inwieweit Algorithmen bei der Erreichung der Gleichstellungsziele hilfreich oder hinderlich sind, insbesondere unter Berücksichtigung der Phänomene des Gender Bias, Gender Data Gap und des Machine Bias. Am Beispiel von Rekrutierungsalgorithmen werden insbesondere die negativen Konsequenzen für die Gleichstellung von Männern und Frauen erläutert. Der Beitrag zeigt aber auch auf, inwieweit Algorithmen gezielt für die Erreichung von Gleichstellungszielen verwendet werden könnten, unter anderem zur Verfolgung positiver Maßnahmen und zur Aufdeckung von Diskriminierungen. The article analyses to what extent algorithms enable or hinder the achievement of gender equality goals, especially considering the phenomena of gender bias, machine bias and the gender data gap. The, mostly negative, consequences for gender equality will be explained drawing on the example of recruitment algorithms. However, it will also be shown to what extent algorithms could be used specifically to achieve gender equality, for instance by using algorithms in positive action measures and to detect discrimination.

We study the problem of sub-trajectory nearest-neighbor queries on polygonal curves under the continuous Fréchet distance. Given an n vertex trajectory P and an m vertex query trajectory Q , we seek to report a vertex-aligned sub-trajectory P ′ of P that is closest to Q , i.e., P′ must start and end on contiguous vertices of P . Since in real data P typically contains a very large number of vertices, we focus on answering queries, without restrictions on P or Q , using only precomputed structures of 𝒪(n) size. We use three baseline algorithms from straightforward extensions of known work; however, they have impractical performance on realistic inputs. Therefore, we propose a new Hierarchical Simplification Tree (HST) data structure and an adaptive clustering-based query algorithm that efficiently explores relevant parts of P . The core of our query methods is a novel greedy-backtracking algorithm that solves the Fréchet decision problem using 𝒪(n+m) space and 𝒪O(nm) time in the worst case. Experiments on real and synthetic data show that our heuristic effectively prunes the search space and greatly reduces computations compared to baseline approaches.

AI edge devices require local intelligence for the concerns of latency and privacy. Given the accuracy and energy constraints, low-power convolutional neural networks (CNNs) are gaining popularity. To alleviate the high memory access energy and computational cost of large CNN models, prior works have proposed promising approaches including in-memory-computing (IMC) [1], mixed-signal multiply-and-accumulate (MAC) calculation [2], and reduced resolution network –[4]. With weights and activations restricted to ±1, binary neural network (BNN) combining with IMC greatly improves the storage and computation efficiency, making it well-suited for edge-based applications, and has demonstrated state-of-the-art energy efficiency in image classification problems [5]. However, compared to full resolution network, BNN requires larger model thus more operations (OPs) per inference for a certain accuracy. To address such challenge, we propose a mixed-signal ternary CNN based processor featuring higher energy efficiency than BNN. It confers several key improvements: 1) the proposed ternary network provides 1.5-b resolution (0/+1/-1), leading to 3.9x OPs/inference reduction than BNN for the same MNIST accuracy; 2) a 1.5b MAC is implemented by V CM -based capacitor switching scheme, which inherently benefits from the reduced signal swing on the capacitive DAC (CDAC); 3) the V CM -based MAC introduces sparsity during training, resulting in lower switching rate. With a complete neural network on chip, the proposed design realizes 97.1% MNIST accuracy with only 0.18uJ per classification, presenting the highest power efficiency for comparable MNIST accuracy.

This study aims at designing a robust nonparametric identifier for a class of singular perturbed systems (SPSs) with uncertain mathematical models. The identifier structure uses a novel identifier based on a differential neural network (DNN) with rational form, which can take into account the multirate nature of SPS. The identifier uses a mixed learning law including a rational formulation of neural networks which is useful to solve the identification of the fast dynamics in the SPS dynamics. The rational form of the design is proposed in such a way that no-singularities (denominator part of the rational form never touches the origin) are allowed in the identifier dynamics. A proposed control Lyapunov function and a nonlinear parameter identification methodology yield to design the learning laws for the class of novel rational DNN which appears as the main contribution of this study. A complementary matrix inequality-based optimization method allows to get the smallest attainable convergence invariant region. A detailed implementation methodology is also given in the study with the aim of clarifying how the proposed identifier can be used in diverse SPSs. A numerical example considering the dynamics of the enzymatic-substrate-inhibitor system with uncertain dynamics is showing how to apply the DNN identifier using the multirate nature of the proposed DNN identifier for SPSs. The proposed identifier is compared to a classical identifier which is not taking into account the multirate nature of SPS. The benefits of using the rational form for the identifier are highlighted in the numerical performance comparison based on the mean square error (MSE). This example justifies the ability of the suggested identifier to reconstruct both the fast and slow dynamics of the SPS.

This paper investigates the elliptic localization for moving object problem from time delay (TD) and Doppler frequency shift (DFS) measurements, where the transmitter position and velocity are unknown. The transmitter is not perfectly time syncronized such that unknown offsets exist in the TD and DFS measurements. We propose to jointly estimate the object and transmitter positions and velocities and the offsets. Using the TD and DFS measurements from both the indirect and direct paths between the transmitter and the receivers, we formulate a non-convex weighted least squares (WLS) problem. Local convergence may occur when solving the non-convex WLS problem, implying that good estimate is not guaranteed. Thus, we relax the non-convex WLS problem into a convex semidefinite program by applying semidefinite relaxation (SDR). Moreover, we theoretically show that the performance can be improved by using multiple transmitters as compared to that using single transmitter, although more unknown parameters are introduced. We then extend the proposed SDR method to handle the multiple transmitters case. Finally, the mean square error analysis is provided to show that the proposed WLS method reaches the Cramer-Rao lower bound accuracy under small Gaussian noise condition. Simulation results validate the theoretical analysis and show the superior performance over the existing methods.

Degree-of-freedom (DoF)-based models have been proven to be highly successful in modeling and analysis of MIMO systems. Among existing DoF-based models, the number of DoFs used for interference cancellation (IC) is solely based on the number of interfering data streams. However, from both experimental and simulation results, we find that signal strengths of an interference link vary significantly in different directions in the eigenspace. In this paper, we exploit the difference in interference signal strengths in the eigenspace and perform IC with DoFs only on those directions with strong signals. To differentiate interference signal strengths on an interference link, we introduce a novel concept called ‘`effective rank threshold.’' Based on this threshold, DoFs are consumed only to cancel strong interferences in the eigenspace while weak interferences are treated as noise in throughput calculation. To better understand the benefits of this approach, we study a fundamental trade-off between network throughput and effective rank threshold for an MU-MIMO network. Our simulation results show that network throughput under optimal rank threshold is significantly higher than that under existing DoF IC models. To ensure the new DoF IC model is feasible at PHY layer, we propose an algorithm to set the weights for all nodes that can offer our desired DoF allocation.

Sampling from large dataset is commonly used in the frequent patterns (FPs) mining. To tightly and theoretically guarantee the quality of the FPs obtained from samples, current methods theoretically stabilize the supports of all the patterns in random samples, despite only FPs do matter, so they always overestimate the sample size. We propose an algorithm called multiple sampling-based FPs mining (MSFP). The MSFP first generates the set of approximate frequent items (AFI), and uses the AFI to form the set of approximate FPs without supports ( AFP*), where it does not stabilize the value of any item's or pattern's support, but only stabilizes the relationship ≥ or < between the support and the minimum support, so the MSFP can use small samples to successively obtain the AFI and AFP*, and can successively prune the patterns not contained by the AFI and not in the AFP*. Then, the MSFP introduces the Bayesian statistics to only stabilize the values of supports of AFP*'s patterns. If a pattern's support in the original dataset is unknown, the MSFP regards it as random, and keeps updating its distribution by its approximations obtained from the samples taken in the progressive sampling, so the error probability can be bound better. Furthermore, to reduce the I/O processes in the progressive sampling, the MSFP stores a large enough random sample in memory in advance. The experiments show that the MSFP is reliable and efficient.

Tensor-ring (TR) decomposition was recently studied and applied for low-rank tensor completion due to its powerful representation ability of high-order tensors. However, most of the existing TR-based methods tend to suffer from deterioration when the selected rank is larger than the true one. To address this issue, this article proposes a new low-rank sparse TR completion method by imposing the Frobenius norm regularization on its latent space. Specifically, we theoretically establish that the proposed method is capable of exploiting the low rankness and Kronecker-basis-representation (KBR)-based sparsity of the target tensor using the Frobenius norm of latent TR-cores. We optimize the proposed TR completion by block coordinate descent (BCD) algorithm and design a modified TR decomposition for the initialization of this algorithm. Extensive experimental results on synthetic data and visual data have demonstrated that the proposed method is able to achieve better results compared to the conventional TR-based completion methods and other state-of-the-art methods and, meanwhile, is quite robust even if the selected TR-rank increases.

Since the advent and usage of artificial intelligence approaches in architecture, a significant number of studies have focused on integrating technological solutions to architectural issues. Artificial intelligence applications in architectural design range from intelligent material design to architectural plan solutions. The ubiquity and distribution of research in this field, as well as the rising use of artificial intelligence techniques to solve design challenges, require an analytical classification of the essential literature review. This article presents a descriptive and analytical review of the work on artificial intelligence applications in architecture. A strong review has been made that identifies and addresses the gaps in artificial intelligence and architecture; and the literature review is transformed into statistical plots. The study's findings indicate a growing interest in artificial intelligence in the field of architecture. There is a need for novel research to be conducted in these areas using advanced technology and techniques.

Sorting algorithms plays an important role in the computer science field. Many applications use sorting algorithm. There are several sorting algorithms proposed by experts, namely bubble sort, exchange short, insertion short, heap sort, quick short, merge sort, standard selection sort. One well-known algorithm of sorting is selection sort. In this journal, discussion about standard selection sort is given with thorough analysis. Sorting is very important data structure concepts that has an important role in memory management, file management, in computer science in general, and in many real-life applications. Different sorting algorithms have differences in terms of time complexity, memory use, efficiency, and other factors. There are many sorting algorithms exist right now in the computer science field. Each algorithm has its benefits and limitations where a trade-off exists between execution time and the nature of the complexity of the algorithm itself. The method is theoretical analysis. Three theoretical analyses are given with deep explanation and analysis. Each with six index arrays, namely with six data on it. The numbers are sorted in ascending order. Pseudo code is also given, to understand this algorithm more thoroughly. It is concluded that this theoretical analysis explained the algorithm more clearly, by using process iteration by hand.

In nonstationary environments, data distributions can change over time. This phenomenon is known as concept drift, and the related models need to adapt if they are to remain accurate. With gradient boosting (GB) ensemble models, selecting which weak learners to keep/prune to maintain model accuracy under concept drift is nontrivial research. Unlike existing models such as AdaBoost, which can directly compare weak learners' performance by their accuracy (a metric between [0, 1]), in GB, weak learners' performance is measured with different scales. To address the performance measurement scaling issue, we propose a novel criterion to evaluate weak learners in GB models, called the loss improvement ratio (LIR). Based on LIR, we develop two pruning strategies: 1) naive pruning (NP), which simply deletes all learners with increasing loss and 2) statistical pruning (SP), which removes learners if their loss increase meets a significance threshold. We also devise a scheme to dynamically switch between NP and SP to achieve the best performance. We implement the scheme as a concept drift learning algorithm, called evolving gradient boost (LIR-eGB). On average, LIR-eGB delivered the best performance against state-of-the-art methods on both stationary and nonstationary data.

Choosing a decomposition method for multi-class classification is an important trade-off between efficiency and predictive accuracy. Trying all the decomposition methods to find the best one is too time-consuming for many applications, while choosing the wrong one may result in large loss on predictive accuracy. In this paper, we propose an automatic decomposition method selection approach called ``D-Chooser", which is lightweight and can choose the best decomposition method accurately. D-Chooser is equipped with our proposed difficulty index which consists of sub-metrics including distribution divergence, overlapping regions, unevenness degree and relative size of the solution space. The difficulty index has two intriguing properties: 1) fast to compute and 2) measuring multi-class problems comprehensively. Extensive experiments on real-world multi-class problems show that D-Chooser achieves an accuracy of 83.3% in choosing the best decomposition method. It can choose the best method in just a few seconds, while existing approaches verify the effectiveness of a decomposition method often takes a few hours. We also provide case studies on Kaggle competitions and the results confirm that D-Chooser is able to choose a better decomposition method than the winning solutions.

Time series anomaly detection (TSAD) is an essential problem faced in several fields, e.g. fault detection, fraud detection, and intrusion detection, etc. Although TSAD is a crucial problem in anomaly detection, few solutions in anomaly detection are suitable for it at present. Recently, some researchers use GAN-based methods such as TAnoGAN and TadGAN to solve TSAD problem. However, problems such as model collapse, low generalization capability and poor accuracy still exist. In this paper, we proposed a Dilated Convolutional Transformer-based GAN (DCT-GAN) to enhance accuracy and improve generalization capability of the model. Specifically, DCT-GAN utilize several generators and a single discriminator to alleviate the mode collapse problem. Each generator consists of a dilated convolutional neural network and a Transformer block to obtain fine-grained and coarse-grained information of the time series, which is a useful component to improve generalization capability. We also use weight-based mechanism to balance these generators. Experiments verify the effectiveness of our method and each part of DCT-GAN.

The appearance of Mobile Edge Computing (MEC) successfully solves the bottlenecks of traditional Cloud based networks. Since mobile edges, e.g. base stations, and mobile devices have certain data processing capabilities, it is not necessary to offload all the tasks to the cloud for handling. Therefore, it is quite important to decide the optimal task assignment in MEC systems, and a series of algorithms have been proposed. However, the existing algorithms ignored the data distribution during task assignment, so that their applied ranges are quite limit. Considering the data sharing is quite important in a MEC system, this paper studies task assignment algorithms in Data Shared Mobile Edge Computing Systems in detail. Specifically, three algorithms are proposed to deal with the unordered and ordered holistic tasks respectively. Meanwhile, the situation that the tasks are divisible is also considered, and two algorithms for rearranging the divisible tasks are proposed for different optimization goals. The hardness of the problem, the correctness, complexities, and ratio bounds of the proposed algorithms are analyzed theoretically. Finally, extensive experimental results are carried out. Both theoretical analysis and experiment results show that all the proposed algorithms have high performance in terms of latency, satisfied rate, and energy consumption.

Generative adversarial networks (GANs) are a hot research topic recently. GANs have been widely studied since 2014, and a large number of algorithms have been proposed. However, there is few comprehensive study explaining the connections among different GANs variants, and how they have evolved. In this paper, we attempt to provide a review on various GANs methods from the perspectives of algorithms, theory, and applications. Firstly, the motivations, mathematical representations, and structure of most GANs algorithms are introduced in details. Furthermore, GANs have been combined with other machine learning algorithms for specific applications, such as semi-supervised learning, transfer learning, and reinforcement learning. This paper compares the commonalities and differences of these GANs methods. Secondly, theoretical issues related to GANs are investigated. Thirdly, typical applications of GANs in image processing and computer vision, natural language processing, music, speech and audio, medical field, and data science are illustrated. Finally, the future open research problems for GANs are pointed out.

Several measurement models have been proposed for data from the continuous-reproduction paradigm for studying visual working memory: The original mixture model (Zhang & Luck, 2008) and its extension (Bays, Catalao, & Husain, 2009); the interference measurement model (Oberauer, Stoneking, Wabersich, & Lin, 2017), and the target confusability competition model (Schurgin, Wixted, & Brady, 2020). This article describes a space of possible measurement models in which all existing models can be placed. The space is defined by three dimensions: (1) The choice of a activation function (von-Mises or Laplace), the choice of a response-selection function (variants of Luce’s choice rule or of signal detection theory), and whether or not memory precision is assumed to be a constant over manipulations affecting memory. A factorial combination of these three variables generates all possible models in the model space. Fitting all models to eight data sets revealed a new model as empirically most adequate, which combines a von-Mises activation function with a signal-detection response-selection rule. The precision parameter can be treated as a constant across many experimental manipulations, though it might vary with manipulations not yet explored. All modelling code and the raw data modelled are available on the OSF: osf.io/zwprv

Differentiable ARchiTecture Search (DARTS) uses a continuous relaxation of network representation and dramatically accelerates Neural Architecture Search (NAS) by almost thousands of times in GPU-day. However, the searching process of DARTS is unstable, which suffers severe degradation when training epochs become large, thus limiting its application. In this article, we claim that this degradation issue is caused by the imbalanced norms between different nodes and the highly correlated outputs from various operations. We then propose an improved version of DARTS, namely iDARTS, to deal with the two problems. In the training phase, it introduces node normalization to maintain the norm balance. In the discretization phase, the continuous architecture is approximated based on the similarity between the outputs of the node and the decorrelated operations rather than the values of the architecture parameters. Extensive evaluation is conducted on CIFAR-10 and ImageNet, and the error rates of 2.25% and 24.7% are reported within 0.2 and 1.9 GPU-day for architecture search, respectively, which shows its effectiveness. Additional analysis also reveals that iDARTS has the advantage in robustness and generalization over other DARTS-based counterparts.

A fast non-negative latent factor (FNLF) model for a high-dimensional and sparse (HiDS) matrix adopts a Single Latent Factor-dependent, Non-negative, Multiplicative and Momentum-incorporated Update (SLF-NM2U) algorithm, which enables its fast convergence. It is crucial to achieve a rigorously theoretical proof regarding its fast convergence, which has not been provided in prior research. Aiming at addressing this critical issue, this work theoretically proves that with an appropriately chosen momentum coefficient, SLF-NM2U enables the fast convergence of an FNLF model in both continuous and discrete time cases. Empirical analysis of HiDS matrices generated by representative industrial applications provides empirical evidences for the theoretical proof. Hence, this study represents an important milestone in the field of HiDS matrix analysis.

This study proposed a novel framework for plant leaf disease identification. The proposed model consists of four steps including pre-processing, segmentation, feature extraction, and classification. At first, the unwanted noise and overfitting are removed, and also image contrast level is enhanced. Secondly, the Fuzzy C-Means (FCM) based Chameleon Swarm Algorithm (CSA) named as (FCM-CSA) is used for plant leaf diseased part segmentation. In the third stage, the feature extraction is performed using a fast GLCM feature extraction model. Finally, the Progressive Neural Architecture Search (PNAS) is used for plant leaf disease identification. The experimental investigations are carried out using MATLAB software with the Mendeley database. From this dataset, we have used Apple Cedar Apple Rust (ACAR), Cherry Powdery Mildew (CPM), Corn Common Rust (CCCR), Apple Healthy (AH), Grape Black Rot (GBR), Pepper Bell Bacterial Spot (PBBS), Potato Late Blight (PLB) and Tomato Leaf Mold (TLM) disease images. Different measures such as precision, recall, sensitivity, specificity, and accuracy results are used to validate the performance of the proposed model.

In this article, two novel distributed variational Bayesian (VB) algorithms for a general class of conjugate-exponential models are proposed over synchronous and asynchronous sensor networks. First, we design a penalty-based distributed VB (PB-DVB) algorithm for synchronous networks, where a penalty function based on the Kullback-Leibler (KL) divergence is introduced to penalize the difference of posterior distributions between nodes. Then, a token-passing-based distributed VB (TPB-DVB) algorithm is developed for asynchronous networks by borrowing the token-passing approach and the stochastic variational inference. Finally, applications of the proposed algorithm on the Gaussian mixture model (GMM) are exhibited. Simulation results show that the PB-DVB algorithm has good performance in the aspects of estimation/inference ability, robustness against initialization, and convergence speed, and the TPB-DVB algorithm is superior to existing token-passing-based distributed clustering algorithms.

Deep learning techniques have proven effective in many applications, but these implementations mostly apply to data in one or two dimensions. Handling 3D data is more challenging due to its irregularity and complexity, and there is a growing interest in adapting deep learning techniques to the 3D domain. A recent successful approach called MeshCNN consists of a set of convolutional and pooling operators applied to the edges of triangular meshes. While this approach produced superb results in classification and segmentation of 3D shapes, it can only be applied to edges of a mesh, which can constitute a disadvantage for applications where the focuses are other primitives of the mesh. In this study, we propose face-based and vertex-based operators for mesh convolutional networks. We design two novel architectures based on the MeshCNN network that can operate on faces and vertices of a mesh, respectively. We demonstrate that the proposed face-based architecture outperforms the original MeshCNN implementation in mesh classification and mesh segmentation, setting the new state of the art on benchmark datasets. In addition, we extend the vertex-based operator to fit in the Point2Mesh model for mesh reconstruction from clean, noisy, and incomplete point clouds. While no statistically significant performance improvements are observed, the model training and inference time are reduced by the proposed approach by 91% and 20%, respectively, as compared with the original Point2Mesh model.

Ultrasonic signal acquisition platforms generate considerable amounts of data to be stored and processed, especially when multichannel scanning or beamforming is employed. Reducing the mass storage and allowing high-speed data transmissions necessitate the compression of ultrasonic data into a representation with fewer bits. High compression accuracy is crucial in many applications, such as ultrasonic medical imaging and nondestructive testing (NDT). In this study, we present learning models for massive ultrasonic data compression on the order of megabytes. A common and highly efficient compression method for ultrasonic data is signal decomposition and subband elimination using wavelet packet transformation (WPT). We designed an algorithm for finding the wavelet kernel that provides maximum energy compaction and the optimal subband decomposition tree structure for a given ultrasonic signal. Furthermore, the WPT convolutional autoencoder (WPTCAE) compression algorithm is proposed based on the WPT compression tree structure and the use of machine learning for estimating the optimal kernel. To further improve the compression accuracy, an autoencoder (AE) is incorporated into the WPTCAE model to build a hybrid model. The performance of the WPTCAE compression model is examined and benchmarked against other compression algorithms using ultrasonic radio frequency (RF) datasets acquired in NDT and medical imaging applications. The experimental results clearly show that the WPTCAE compression model provides improved compression ratios while maintaining high signal fidelity. The proposed learning models can achieve a compression accuracy of 98% by using only 6% of the original data.

The excellent remote sensing ability of synthetic aperture radar (SAR) will be misled seriously when it encounters deceptive jamming which possesses high fidelity and fraudulence. In this paper, the dynamic synthetic aperture (DSA) scheme is used to extract the difference between the true and false targets. A simultaneous deceptive jamming suppression and target reconstruction method is proposed for a single channel SAR system to guarantee remote sensing ability. The system model is formulated as a sparse signal recovery problem with an unknown parametric dictionary to be estimated. An iterative re-weighted method is employed to jointly handle the dictionary parameter learning and target reconstruction problem in an majorization-minimization framework, where a surrogate function majorizing the Gaussian entropy in the objective function is introduced to circumvent its non-convexity. After dictionary parameter learning, the grid mismatching problem in a fixed grid based method is avoided. Therefore, the proposed method can reap a super resolution result. Besides, a simple yet effective DSA section scheme is developed for the SAR data excerpting, in which only two DSAs are required. Experimental results about location error and reconstruction power error reveal that the proposed method is able to achieve a good performance in deceptive jamming suppression.

Electromyostimulation is a method of restorative treatment based on electrical stimulation of nerves and muscles. The electric current, which is used in electrical stimulation to obtain induced muscle contractions, is characterized by a large number of different parameters. However, not every possible option of electrical stimulation is highly effective. To solve the task of diagnosing some diseases of the neuromuscular system, it is important to organize the software by analyzing the parameters of the evoked potentials. Therefore, the object of research is the processes of skeletal muscle contraction under the influence of natural electrical pulses of the nervous system or under the influence of external electrical stimulation. The subject of research is models describing the processes in muscles during contraction and methods of data processing. In the course of the study, such research methods as mathematical modeling methods and methods of processing medical and biological data were used. The paper examines the experimental strength-duration dependence of skeletal muscle and obtained mathematical models for the normal state of the neuromuscular apparatus and different degrees of denervation. On the basis of electrodiagnosis of a patient with impaired motor functions, the dynamics of changes in the patient's condition and the effectiveness of treatment were traced. Based on the results of the study, an analysis of the parameters of the evoked potentials of the stimulation electromyogram during adaptive electrostimulation was carried out in order to control its effectiveness or establish a diagnosis in some diseases of the neuromuscular system. This made it possible to develop a method for correcting errors in the interpretation of one of the quality parameters and increase the reliability of the diagnosis. The obtained results can be used in the improvement of technical devices for electrostimulation therapy, as well as control of the effectiveness of rehabilitation procedures.

With the rapid development of Internet of Things (IoT) and next-generation communication technologies, resource-constrained mobile devices fail to meet the demand of resource-hungry and compute-intensive applications. To cope with this challenge, with the assistance of Mobile Edge Computing (MEC), offloading complex tasks from mobile devices to edge cloud servers or central cloud servers can reduce the computational burden of devices and improve the efficiency of task processing. However, it is difficult to obtain optimal offloading decisions by conventional heuristic optimization methods, because the decision-making problem is usually NP-hard. In addition, there are shortcomings in using intelligent decision-making methods, e.g., lack of training samples and poor ability of migration under different MEC environments. To this end, we propose a novel offloading algorithm named MR-DRO, consisting of a Meta-Reinforcement Learning (meta-RL) model, which improves the migration ability of the whole model, and a Deep Reinforcement Learning (DRL) model, which combines multiple parallel Deep Neural Networks (DNNs) to learn from historical task offloading scenarios. Simulation results demonstrate that our approach can effectively and efficiently generate near-optimal offloading decisions in IoT environments with edge and cloud collaboration, which further improves the computational performance and has strong portability when making offloading decisions.

Despite the empirical success in various domains, it has been revealed that deep neural networks are vulnerable to maliciously perturbed input data that can dramatically degrade their performance. These are known as adversarial attacks. To counter adversarial attacks, adversarial training formulated as a form of robust optimization has been demonstrated to be effective. However, conducting adversarial training brings much computational overhead compared with standard training. In order to reduce the computational cost, we propose an annealing mechanism, annealing mechanism for adversarial training acceleration (Amata), to reduce the overhead associated with adversarial training. The proposed Amata is provably convergent, well-motivated from the lens of optimal control theory, and can be combined with existing acceleration methods to further enhance performance. It is demonstrated that, on standard datasets, Amata can achieve similar or better robustness with around 1/3-1/2 the computational time compared with traditional methods. In addition, Amata can be incorporated into other adversarial training acceleration algorithms (e.g., YOPO, Free, Fast, and ATTA), which leads to a further reduction in computational time on large-scale problems.

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

Due to the unavoidable influence of sparse and Gaussian noise during the process of data acquisition, the quality of hyperspectral images (HSIs) is degraded and their applications are greatly limited. It is therefore necessary to restore clean HSIs. In the traditional methods, low-rank and sparse matrix decomposition methods are usually applied to restore the pure data matrix from the observed data matrix. However, due to the fact that the optimization of the l0-norm for the sparse modeling is a non-convex and NP-hard problem, convex relaxation and regularization parameters are usually introduced. However, convex relaxation often leads to inaccurate sparse modeling results, and the sensitive regularization parameters can lead to unstable results. Thus, in this paper, to address these issues, an accurate multi-objective low-rank and sparse denoising framework is proposed for HSIs to achieve accurate modeling. The l0-norm is directly modeled as the sparse noise and is optimized by an evolutionary algorithm, and the denoising problem is converted into a multi-objective optimization problem through simultaneously optimizing the low-rank term, the sparse term, and the data fidelity term, without sensitive regularization parameters. However, since the low-rank clean image and sparse noise of the HSI are encoded into a solution, the length of the solution is too long to be optimized. In this paper, a subfitness strategy is constructed to achieve effective optimization through comparing the objective function values corresponding to each band for each solution. The experiments undertaken with simulated images in 11 noise cases and four real noisy images confirm the effectiveness of the proposed method.

We consider a generalization of the recently proposed gradient coding framework where a large dataset is divided across $n$ workers and each worker transmits to a master node one or more linear combinations of the gradients over the data subsets assigned to it. Unlike the conventional framework which requires the master node to recover the sum of the gradients over all the data subsets in the presence of $s$ straggler workers, we relax the goal of the master node to computing the sum of at least some α fraction of the gradients. The broad goal of our work is to study the optimal computation and communication load per worker for this approximate gradient coding framework. We begin by deriving a lower bound on the computation load of any feasible scheme and also propose a strategy which achieves this lower bound, albeit at the cost of high communication load and a number of data partitions which can be polynomial in the number of workers n. We then restrict attention to schemes which utilize a number of data partitions equal to $n$ and propose schemes based on cyclic assignment which have a lower communication load. When each worker transmits a single linear combination, we also prove lower bounds on the computation load of any scheme using $n$ data partitions. A full version of this paper is accessible at: https://arxiv.org/abs/2102.10163

This paper evaluates outage probability (OP) and intercept probability (IP) of physical-layer security based MIMO networks adopting cooperative jamming (Coop-Jam). In the considered scenario, a multi-antenna source communicates with a multi-antenna destination employing transmit antenna selection (TAS)/ selection combining (SC), in presence of a multi-antenna eavesdropper using SC. One of jammers appearing near the destination is selected for generating jamming noises on the eavesdropper. Moreover, the destination supports the wireless energy for the chosen jammer, and cooperates with it to remove the jamming noises. We consider two jammer selection approaches, named RAND and SHORT. In RAND, the destination randomly selects the jammer, and in SHORT, the jammer, which is nearest to the destination, is chosen. We derive exact and asymptotic expressions of OP and IP over Rayleigh fading, and perform Monte-Carlo simulations to verify the correction of our derivation. The results present advantages of the proposed RAND and SHORT methods, as compared with the corresponding one without using Coop-Jam.

With the growth of the Internet of Vehicles (IoV) in Intelligent Autonomous Transport Systems (IATS), a huge volume of data is exchanged between vehicles in these newly developed infrastructures. As a result, the requirements of securing data exchange between vehicles, autonomous or otherwise, have also increased tremendously. Securing data transfer and keeping a record of each transaction becomes a necessity in IoV/IATS. In this paper, we propose some optimized security algorithms using symmetric encryption for secure multimedia data transfer between vehicles. The main feature of these optimized algorithms is that they use a lower amount of data to generate fingerprints. The algorithms convert approximately 3.7 x 10⁵ samples of data into 3600 samples to generate the fingerprint. Fast Fourier Transform (FFT) is used to fetch the highest three peak values of the signal in the frequency domain. A centralized server authenticates the data transfer by comparing the HASH of the fingerprints and also keeps the transaction record. Through experimental analysis, the performance of proposed algorithms is confirmed by achieving reduced size samples to generate fingerprints and their authentication at the server-side.

Identifying geometric features from sampled surfaces is a significant and fundamental task. The existing curvature-based methods that can identify ridge and valley features are generally sensitive to noise. Without requiring high-order differential operators, most statistics-based methods sacrifice certain extents of the feature descriptive powers in exchange for robustness. However, neither of these types of methods can treat the surface boundary features simultaneously. In this paper, we propose a novel neighbor reweighted local centroid (NRLC) computational algorithm to identify geometric features for point cloud models. It constructs a feature descriptor for the considered point via decomposing each of its neighboring vectors into two orthogonal directions. A neighboring vector starts from the considered point and ends with the corresponding neighbor. The decomposed neighboring vectors are then accumulated with different weights to generate the NRLC. With the defined NRLC, we design a probability set for each candidate feature point so that the convex, concave and surface boundary points can be recognized concurrently. In addition, we introduce a pair of feature operators, including assimilation and dissimilation, to further strengthen the identified geometric features. Finally, we test NRLC on a large body of point cloud models derived from different data sources. Several groups of the comparison experiments are conducted, and the results verify the validity and efficiency of our NRLC method.

Applying chaos theory for secure digital communications is promising and it is well acknowledged that in such applications the underlying chaotic systems should be carefully chosen. However, the requirements imposed on the chaotic systems are usually heuristic, without theoretic guarantee for the resultant communication scheme. Among all the primitives for secure communications, it is well-accepted that (pseudo) random numbers are most essential. Taking the well-studied two-dimensional coupled map lattice (2D CML) as an example, this paper performs a theoretical study towards pseudo-random number generation with the 2D CML. In so doing, an analytical expression of the Lyapunov exponent (LE) spectrum of the 2D CML is first derived. Using the LEs, one can configure system parameters to ensure the 2D CML only exhibits complex dynamic behavior, and then collect pseudo-random numbers from the system orbits. Moreover, based on the observation that least significant bit distributes more evenly in the (pseudo) random distribution, an extraction algorithm E is developed with the property that, when applied to the orbits of the 2D CML, it can squeeze uniform bits. In implementation, if fixed-point arithmetic is used in binary format with a precision of $z$ bits after the radix point, E can ensure that the deviation of the squeezed bits is bounded by $2^{-z}$ . Further simulation results demonstrate that the new method not only guide the 2D CML model to exhibit complex dynamic behavior, but also generate uniformly distributed independent bits. In particular, the squeezed pseudo random bits can pass both NIST 800-22 and TestU01 test suites in various settings. This study thereby provides a theoretical basis for effectively applying the 2D CML to secure communications., Comment: 10 pages, 11figures

In todays wireless networks, a variety of Radio Access Technologies (RATs) are present. However, each RAT being controlled individually leads to suboptimal utilization of network resources. Due to the remarkable growth of data traffic, interworking among different RATs is becoming necessary to overcome the problem of suboptimal resource utilization. Users can be offloaded from one RAT to another based on loads of different networks, channel conditions and priority of users. We consider the optimal RAT selection problem in a Fifth Generation (5G) New Radio (NR)-Wireless Fidelity (WiFi) network where we aim to maximize the total system throughput subject to constraints on the blocking probability of high priority users and the offloading probability of low priority users. The problem is formulated as a Constrained Markov Decision Process (CMDP). We reduce the effective dimensionality of the action space by eliminating the provably suboptimal actions. We propose low-complexity online heuristics for RAT selection which can operate without the knowledge regarding the statistics of system dynamics. Network Simulator-3 (ns-3) simulations reveal that the proposed algorithms offer near-optimal performances and outperform traditional RAT selection algorithms under realistic network scenarios including user mobility.

Finding optimal parameters for acceleration structures for raytracing is key to improved performance. Previous research has shown that a speedup of over 10% of rendering time is possible. Some parameters are interdependent which complicates the process of finding an optimal configuration. It is hence interesting to find them efficiently. Autotuning is an automatic optimization scheme able to search for optimal configurations and has been applied successfully to kD-trees in the past, which we apply today on BVHs. The more parameters to optimize, the more difficult it is to find optimal solutions. In this paper, we analyze in detail the behavior of the parameters and their impact on acceleration structure building and rendering time. We show the interdependence and context sensitivity (i.e., scene, viewpoint) of the parameters. Based on the use case, this allows to target only crucial parameters. Convergence speed towards an optimal configuration is essential. To find better parameters, the autotuner needs to build the acceleration structure over and over, changing parameters every time. We introduce a hybrid model-based prediction and online autotuning method to address this issue. The prediction model allows for both instantaneous near-optimal configurations when inputs are known or similar, and efficient search of the configuration space when inputs are completely new. Online autotuning outperforms configurations recommended in literature by up to 11% median. The prediction model achieves 95% of the maximum speedup of the autotuner while reducing 90% of its overhead. Thus, hybrid online autonuning enables always-on tuning in ray tracing.

Relating neural activity to behavior requires an understanding of how neural computations arise from the coordinated dynamics of distributed, recurrently connected neural populations. However, inferring the nature of recurrent dynamics from partial recordings of a neural circuit presents significant challenges. Here, we show that some of these challenges can be overcome by a fine-grained analysis of the dynamics of neural residuals, i.e. trial-by-trial variability around the mean neural population trajectory for a given task condition. Residual dynamics in macaque pre-frontal cortex (PFC) in a saccade-based perceptual decision-making task reveals recurrent dynamics that is time-dependent, but consistently stable, and suggests that pronounced rotational structure in PFC trajectories during saccades is driven by inputs from upstream areas. The properties of residual dynamics restrict the possible contributions of PFC to decision-making and saccade generation, and suggest a path towards fully characterizing distributed neural computations with large-scale neural recordings and targeted causal perturbations.

The goal of multi-view clustering is to partition samples into different subsets according to their diverse features. Previous multi-view clustering methods mainly exist two forms: multi-view spectral clustering and multi-view matrix factorization. Although they have shown excellent performance in many occasions, there are still many disadvantages. For example, multi-view spectral clustering usually needs to perform postprocessing. Multi-view matrix factorization directly decomposes the original data features. When the size of features is large, it encounters the expensive time consumption to decompose these data features thoroughly. Therefore, we proposed a novel multi-view clustering approach. The main advantages include the following three aspects: 1) it searches for a common joint graph across multiple views, which fully explores the hidden structure information by utilizing the compatibility among views; 2) the introduced nonnegative constraint manipulates that the final clustering results can be directly obtained; and 3) straightforwardly decomposing the similarity matrix can transform the eigenvalue factorization in spectral clustering with computational complexity O(n³) into the singular value decomposition (SVD) with O(nc²) time cost, where n and c, respectively, denote the numbers of samples and classes. Thus, the computational efficiency can be improved. Moreover, in order to learn a better clustering model, we set that the constructed similarity graph approximates each view affinity graph as close as possible by adding the constraint as the initial affinity matrices own. Furthermore, substantial experiments are conducted, which verifies the superiority of the proposed two clustering methods comparing with single-view clustering approaches and state-of-the-art multi-view clustering methods.

The use of the internet in this modern era is increased many fold. The communications between different peers take place in digital form. While sharing the messages between different recipients, the confidentiality of the messages is very important. For creating the high level of security while sharing the secret messages, the cryptographic algorithms are introduced along with steganography. Image Steganography is a methodology used to hide the messages inside of the cover image. Initially, the secret information is encrypted by using the RSA Algorithm. Then the encrypted secret information is hidden in the Least Significant Bit (LSB) of the different components of the color image in such a way that the original quality of the image to be minimized. The recipient of the message is able to retrieve the encrypted secret message from the LSB bit of stego_image and then the cipher text is converted into original plain text by using the RSA algorithm. The proposed algorithm verified and analysed its performance against the different combinations of key pairs.

The most significant task aimed at maintaining the bridge is the “bridge deck crack inspection”. Conventionally, the inspector identifies the cracks utilizing eyes and notices the crack location manually. Nevertheless, accuracy of inspection outcome is less because of “subjective nature of human judgment”. And Suggested a “crack inspection system” which utilizes the mobile robot equipped with the camera to gather bridge deck pictures. And in this model, “Laplacian of Gaussian (LoG) algorithm” is utilized to identify cracks & “global crack map” is achieved by localization of robot and “camera calibration”. Ensuring that robot gathers entire bridge deck pictures, the “path planning algorithm” on the basis of genetic-algorithm will be improved. And the “path planning algorithm” identifies the solution that reduces the count of turns and moving distance. The evaluation is done on the suggested method by both experiments & simulations.

Nyquist Folding Receiver (NYFR) is a perceptron structure that realizes a low probability of intercept (LPI) signal analog to information. Aiming at the problem of LPI radar signal receiving, the time domain, frequency domain, and time-frequency domain problems of signals intercepted by NYFR structure are studied. Combined with the time-frequency analysis (TFA) method, a radar recognition scheme based on deep learning (DL) is introduced, which can reliably classify common LPI radar signals. First, the structure of NYFR and its characteristics in the time domain, frequency domain, and time and frequency domain are analyzed. Then, the received signal is then converted into a time-frequency image (TFI). Finally, four kinds of DL algorithms are used to classify LPI radar signals. Simulation results demonstrate the correctness of the NYFR structure, and the effectiveness of the proposed recognition method is verified by comparison experiments.

Mobile Ad-hoc Network (MANET) is a multitude of self-organizing, self-configuring and infrastructure-less network of mobile devices. In MANET, the node will be act as a router. Routing is a main challenge to perpetuate connectivity and verify the successive rate of Quality of Service (QoS). In order to provide the accurate data delivery with improved QoS the modified Lumer and Faieta algorithm with modified Ad hoc On-Demand Distance Vector Routing (AODV), Dynamic Source Routing (DSR) reactive protocols are used in the proposed work. The routing process is controlled depending on traffic protocols to minimize the network latency. In addition, QoS provision based on delay requirement to ensure guaranteed performance level to the delay sensitive applications. Three performance parameters such as End-to-end Average Delay, Packet Delivery Ratio (PDR) and Control Overhead are contemplated in different mobility scenarios to compare the performance of the proposed protocol by using NS2 simulator with the original AODV, DSR and the MAC layer information can be used to construct the neighbor table which will enhance the performance in terms of delay and PDR as well as minimizing the control overhead.

Plants make up more than 80 percent of the human diet. As a result, they are critical for food security and ensuring that we all have access to enough, cheap, clean, and nutritional food to lead healthy and active living. This research focuses on plant diseases as they create a major threat to food security. Expert people are hired in traditional farming to physically evaluate line by line for host disease plants, which would be labor-intensive, time taking and potentially error-prone activity because it is done by people manually. This work focus on, supervised machine learning algorithms like RF, SVM, DT, KNN, NB, and KNN with image processing methods and also analysis algorithm results with each other and finds the best algorithm for plant disease classification. RF algorithm achieved 89 percent accuracy compared with other algorithms.

The goal of this experiment is to investigate the effect of laser engraving parameters on Al 6063 aluminum alloy specimens using a fiber laser engraving equipment. The experimental design makes use of Taguchi's L16 orthogonal array. Surface roughness and engraving depth are the responses which are to be optimum for machining process parameters such as fill spacing, power, frequency, and scan speed. Using regression analysis, the mathematical models for individual responses were created. Grey relational analysis coupled with the Taguchi technique has been presented for simultaneous response optimization. The optimal setting parameters for multiple performance characteristics is fill spacing at level-3, power at level-4, frequency at level-2 and scan speed at level-4 i.e., FS3-P4-F2-SS4. To determine the significance of the models, the Analysis of Variance (ANOVA) approach is used. Finally, in the confirmation test, it is observed that the grey relational grade is improved from the initial process parameter combination.

With the tremendous advancement in machine learning and deep learning, organizations are using numerous algorithms for analyzing the huge amount of data to come up with insights which contains meaningful out comes. Especially in medical health care systems, machine learning is being used widely to forecast and treat the illness at early in stage. Cervical cancer is one of such diseases which can be diagnosed if identified in early stage, but the symptom-less nature of the disease has become challenging for the researchers and practitioners to predict the disease at early stage. There are numerous factors which will prone to cervical cancer risk in women. In this study we have proposed a stacked ensemble technique which takes heterogenous base learners and a meta learner for predicting the cervical cancer from various risk factors, SMOTE has been used for data Balancing and RFE with Random Forest used for feature extraction through which we have got better accuracy than the existing methods. We have also identified top 8 features which impact the advancement of the classification model.

The cloud server stores its unique DE-duplicate data replication cloud storage system as outsourced only data copies. Deduplication cloud storage audits can verify the integrity of data stored in the cloud, while the cloud calculates the need to maintain only one copy of the copied file. The existing de-duplication auditing solutions for cloud storage are vulnerable to brute-force dictionary attacks, which is a disclosure leakage of user privacy and more time consumption to find a duplicate files in cloud. Users must check the integrity of lightweight cloud computing data to create data authentication and retrieve files in the cloud. The proposed system uses a secure Array Preserving Triple Data Encryption (AP3DE) algorithm to verify with confidence the expertise of aggregate technology. In this method to analysis the user upload document before store on cloud. The aggregator to provide an authentication access the data access role to cloud user. Cloud protects the privacy, confidentiality, and integrity of data stored in the cloud by generating the code of the authentication file and the encryption key of the file through the proposed system test. The Index-level Boundary Pattern Convergent Encryption (IBPCE) algorithm integrates dynamic data handling and data exclusion capabilities for secure cloud storage privacy security public audits. The cloud data security using a Convergent international data encryption algorithm used to encrypted data. It is a 64 bit plaintext and cipher text blocks is controlled by 128 bit key. Analysis of safety and efficiency shows that the proposed mechanism is highly efficient and safe. In this proposed IBPCE method to manage and reduce the cloud storage compare to other existing method.

BCube is a well-known recursively defined network structure. It provides multiple low-diameter paths and good fault-tolerance for data center networks (DCNs). Its distributed routing algorithm, BCube Source Routing (BSR), can be deployed rapidly and conveniently to build multiple parallel path sets. But in the worst case, BSR may suffer from flow collisions and waste $50\%$ of the capacity of each BCube link. In this paper, to fully use those redundant links and to improve BCube load balancing, we supplement the BCube topology with a central master computer and we design two centralized dynamic parallel flow scheduling algorithms: CDPFS and CDPFSMP, for single-path and multi-path respectively. We focus on finding the least congested path for each flow by analyzing the information about the state of the global network. Furthermore, we allocate those paths to each flow in parallel. The simulation result shows that our proposed algorithms take advantage of BCube structure and deliver high-performance solutions for load balancing problems.