Search

We present a solution to egocentric 3D body pose estimation from monocular images captured from downward looking fish-eye cameras installed on the rim of a head mounted VR device. This unusual viewpoint leads to images with unique visual appearance, with severe self-occlusions and perspective distortions that result in drastic differences in resolution between lower and upper body. We propose an encoder-decoder architecture with a novel multi-branch decoder designed to account for the varying uncertainty in 2D predictions. The quantitative evaluation, on synthetic and real-world datasets, shows that our strategy leads to substantial improvements in accuracy over state of the art egocentric approaches. To tackle the lack of labelled data we also introduced a large photo-realistic synthetic dataset. xR-EgoPose offers high quality renderings of people with diverse skintones, body shapes and clothing, performing a range of actions. Our experiments show that the high variability in our new synthetic training corpus leads to good generalization to real world footage and to state of theart results on real world datasets with ground truth. Moreover, an evaluation on the Human3.6M benchmark shows that the performance of our method is on par with top performing approaches on the more classic problem of 3D human pose from a third person viewpoint., 14 pages. arXiv admin note: substantial text overlap with arXiv:1907.10045

For face-related applications like face recognition and face verification, face detection is vital. Law enforcement agencies extensively use face recognition applications for the detection of known offenders. Security cameras are increasingly deployed everywhere, and these cameras’ captured data can be used for tagging offenders. This is done by offering the face detection system with input images, which are then processed. Later on, attempts will be done to tag the security cameras’ captured data with this processed input data. Effective utilization of deep learning techniques in order to efficiently carry out this task. Computer vision’s various tasks have been taken over by deep Convolutional Neural Networks (CNNs). Of late, the key paradigms for face detection are the region-based CNN (R-CNN) detection models due to their quicker processing speed and progressively better performance. Nevertheless, the huge feature set causes these models’ to be inefficient for real-time face detection. A Grammatical Evolution (GE) Region-Based Fully Conventional Neural Network (R-FCN) for face detection has been proposed in this work in which the GE heuristics are utilized for tuning the hyper-parameters. The WIDER face dataset and Face Detection Dataset and Benchmark (FDDB) dataset were employed for conducting the assessments. In comparison to other advanced paradigms, the proposed technique’s superiority is evident from the experiments conducted.

Deep Neural Network (DNN) in recent era offers new opportunities to manage the production systems with increasing complexity. In this paper, DNN is trained with discrete event simulation and a process based association to make the job shop an autonomous one. This intelligent system operates well in complex environment with constrained time limits while making optimal decisions. The DNN is henceforth combined with constrained time limits for the process of production control. The system is implemented typically in semiconductor manufacturing industries on complex job shops of a wafer fab case. The DNN is trained in such a way that it operates in complex environment with timing constraints that ships the job in accurate way without flaws in operation. The DNN rewards the selection with reduced time constraint, which tends to operates with most critical batch list. The study therefore shows that the DNN manages well the timing constraints than the standard benchmark technique.

In plenty of real-life tasks, strongly supervised information is hard to obtain, such that there is not sufficient high-quality supervision to make traditional learning approaches succeed. Therefore, weakly supervised learning has drawn considerable attention recently. In this paper, we consider the problem of learning from incomplete and inaccurate supervision, where only a limited subset of training data is labeled but potentially with noise. This setting is challenging and of great importance but rarely studied in the literature. We notice that in many applications, the limited labeled data are usually with one-sided noise. For instance, considering the bug detection task in the software system, the identified buggy codes are indeed with defects whereas the codes that have been checked many times or newly fixed may still have other flaws due to the complexity of the system. We propose a novel method which is able to effectively alleviate the negative influence of one-sided label noise with the help of a vast number of unlabeled data. Excess risk analysis is provided as theoretical justifications on the usefulness of incomplete and one-sided inaccurate supervision. We conduct experiments on synthetic, benchmark datasets, and real-life tasks to validate the effectiveness of the proposed approach.

In this article, we propose a hybrid conjugate gradient (CG) scheme for solving unconstrained optimization problem. The search direction is a combination of the Polak–Ribiere–Polyak (PRP) and the Liu–Storey (LS) CG parameters and is close to the direction of the memoryless Broyden–Fletcher–Goldfarb–Shanno (BFGS) quasi-Newton scheme. Without the use of the line search, the search direction satisfies the descent condition and possesses the trust region property. The global convergence of the scheme for general functions under the Wolfe-type and Armijo-type line search is established. Numerical experiments are carried out on some benchmark test problems and the results show that the propose scheme is more efficient than other existing schemes. Finally, a practical application of the scheme in motion control of robot manipulator is also presented.

The incompleteness of knowledge graphs triggers considerable research interest in relation prediction. As the key to predicting relations among entities, many efforts have been devoted to learning the embeddings of entities and relations by incorporating a variety of neighbors' information which includes not only the information from direct outgoing and incoming neighbors but also the ones from the indirect neighbors on the multihop paths. However, previous models usually consider entity paths of limited length or ignore sequential information of the paths. Either simplification will make the model lack a global understanding of knowledge graphs and may result in the loss of important and indispensable information. In this article, we propose a novel global graph attention embedding network (GGAE) for relation prediction by combining global information from both direct neighbors and multihop neighbors. Concretely, given a knowledge graph, we first introduce the path construction algorithms to obtain meaningful paths, then design path modeling methods to capture the potential long-distance sequential information in the multihop paths, final propose an entity graph attention and a relation graph attention mechanisms to obtain entity embeddings and relation embeddings. Moreover, an entity graph attention mechanism is proposed to calculate the entity embeddings by aggregating direct incoming and outgoing neighbors from: 1) an original knowledge graph with the original entity and relation embeddings and 2) a new knowledge graph constructed by the paths whose embeddings are updated by path modeling methods. for each relation, we construct a new graph with related entities and present a relation graph attention to learn the features. Therefore, our model can encapsulate the information from different distance neighbors, and enable the embeddings of entities and relations to better capture all-sided semantic information. The experimental results on benchmark datasets verify the superiority of our model over the state-of-the-art ones.

Since real-world objects and their interactions are often multi-modal and multi-typed, heterogeneous networks have been widely used as a more powerful, realistic, and generic superclass of traditional homogeneous networks (graphs). Meanwhile, representation learning (\aka~embedding) has recently been intensively studied and shown effective for various network mining and analytical tasks. In this work, we aim to provide a unified framework to deeply summarize and evaluate existing research on heterogeneous network embedding (HNE), which includes but goes beyond a normal survey. Since there has already been a broad body of HNE algorithms, as the first contribution of this work, we provide a generic paradigm for the systematic categorization and analysis over the merits of various existing HNE algorithms. Moreover, existing HNE algorithms, though mostly claimed generic, are often evaluated on different datasets. Understandable due to the application favor of HNE, such indirect comparisons largely hinder the proper attribution of improved task performance towards effective data preprocessing and novel technical design, especially considering the various ways possible to construct a heterogeneous network from real-world application data. Therefore, as the second contribution, we create four benchmark datasets with various properties regarding scale, structure, attribute/label availability, and \etc.~from different sources, towards handy and fair evaluations of HNE algorithms. As the third contribution, we carefully refactor and amend the implementations and create friendly interfaces for 13 popular HNE algorithms, and provide all-around comparisons among them over multiple tasks and experimental settings., Accepted by IEEE TKDE. All code and data available at https://github.com/yangji9181/HNE

Constrained multiobjective optimization problems (CMOPs) involve both conflicting objective functions and various constraints. Due to the presence of constraints, CMOPs' Pareto-optimal solutions are very likely lying on constraint boundaries. The experience from the constrained single-objective optimization has shown that to quickly obtain such an optimal solution, the search should surround the boundary of the feasible region from both the feasible and infeasible sides. In this article, we extend this idea to cope with CMOPs and, accordingly, we propose a novel constrained multiobjective evolutionary algorithm with bidirectional coevolution, called BiCo. BiCo maintains two populations, that is: 1) the main population and 2) the archive population. To update the main population, the constraint-domination principle is equipped with an NSGA-II variant to move the population into the feasible region and then to guide the population toward the Pareto front (PF) from the feasible side of the search space. While for updating the archive population, a nondominated sorting procedure and an angle-based selection scheme are conducted in sequence to drive the population toward the PF within the infeasible region while maintaining good diversity. As a result, BiCo can get close to the PF from two complementary directions. In addition, to coordinate the interaction between the main and archive populations, in BiCo, a restricted mating selection mechanism is developed to choose appropriate mating parents. Comprehensive experiments have been conducted on three sets of CMOP benchmark functions and six real-world CMOPs. The experimental results suggest that BiCo can obtain quite competitive performance in comparison to eight state-of-the-art-constrained multiobjective evolutionary optimizers.

Multi-view clustering has attracted increasing attention in multimedia, machine learning and data mining communities. As one kind of the essential multi-view clustering algorithm, multi-view subspace clustering (MVSC) becomes more and more popular due to its strong ability to reveal the intrinsic low dimensional clustering structure hidden across views. Despite superior clustering performance in various applications, we observe that existing MVSC methods directly fuse multi-view information in the similarity level by merging noisy affinity matrices; and isolate the processes of affinity learning, multi-view information fusion and clustering. Both factors may cause insufficient utilization of multi-view information, leading to unsatisfying clustering performance. This paper proposes a novel consensus one-step multi-view subspace clustering (COMVSC) method to address these issues. Instead of directly fusing multiple affinity matrices, COMVSC optimally integrates discriminative partition-level information, which is helpful to eliminate noise among data. Moreover, the affinity matrices, consensus representation and final clustering labels matrix are learned simultaneously in a unified framework. By doing so, the three steps can negotiate with each other to best serve the clustering task, leading to improved performance. Accordingly, we propose an iterative algorithm to solve the resulting optimization problem. Extensive experiment results on benchmark datasets demonstrate the superiority of our method against other state-of-the-art approaches.

Short-range wireless technologies will enable vehicles to communicate and coordinate their actions, thus improving people's safety and traffic efficiency. Whereas IEEE 802.11p (and related standards) had been the only practical solution for years, in 2016 a new option was introduced with Release of 14 long-term evolution (LTE), which includes new features to enable direct vehicle-to-vehicle (V2V) communications. LTE-V2V promises a more efficient use of the channel compared to IEEE 802.11p, thanks to an improved PHY layer and the use of orthogonal resources at the MAC layer. In LTE-V2V, a key role is played by the resource allocation algorithm and increasing efforts are being made to design new solutions to optimize the spatial reuse. In this context, an important aspect still little studied, is therefore that of identifying references that allow, first, to have a perception of the space in which the resource allocation algorithms move and, second, to verify the performance of new proposals. In this paper, we focus on a highway scenario and identify two algorithms to be used as a minimum and maximum reference in terms of the packet reception probability (PRP). The PRP is derived as a function of various parameters that describe the scenario and settings, from the application to the physical layer. Results, obtained both in a simplified Poisson point process scenario and with realistic traffic traces, show that the PRP varies considerably with different algorithms and that there is room for the improvement of current solutions.

In distributed computing, Web Service Composition(WSC)leads to the effective reuse of existing services and products added value. WSC must fulfil functional requirements and optimise the quality of Service (QoS)attributes, simultaneously. Memetic Algorithms (MAs)are promising for automatically com-posing numerous Web services to satisfy the above requirements. Data-intensive Web services focus on providing and updating data with a significant volume of data operation and exchange. However, current composition approaches have ignored the impact of data communication and the distribution of services, which significantly affect the performance when applied to thechallengingDistributed Data-intensive Web Service Composition(DDWSC)problem. Although recent approaches have revealed the usefulness of local search, they have completely overlooked the question of preferring appropriate composition solutions for local search. To address this research issue, we propose a priority-based selection method for the local search that can be consistently integrated with any MA for DDWSC. This enables us to develop state-of-the-art algorithms for DDWSC by explicitly considering the problem-specific, population and solution-related information for choosing a solution. Extensive experimental evaluation using benchmark datasets shows that our proposed method significantly outperforms several recently proposed methods

Software updates are highly significant in autonomous vehicles. These updates are utilized to provide enhanced features and updated security mechanisms. In order to ensure scalability and smooth roll-out Over-the-air (OTA) mechanism is a preferred option to propagate a software update. However, this approach is vulnerable to security attacks because of existence of wireless communication channel between the vehicle and the manufacturer. In that, an attacker can replace the legitimate software with a malicious software with an intent to get control over the vehicle. In this work, we are motivated to address this problem. We develop an enhanced uptane framework for detection of malicious OTA software updates in autonomous vehicles. For enhancing security, we incorporate convolutional neural network (CNN) in the uptane framework. The proposed framework is able to distinguish between malicious and benign software executables with high accuracy. For training and testing, we create two datasets by collecting executables of Windows and Linux operating system. We encourage the use of transfer learning by exploiting the developed CNN models in order to detect malicious executable designed for autonomous vehicles. We also benchmark the CNN models against state-of-the art models. Our work is highly beneficial for the community in providing a secure mechanism for software updates.

In this paper, we introduce, model and solve the pickup and delivery problem with time windows, multiple stacks, and handling operations (PDPTWMS-H). In the PDPTWMS-H, a fleet of vehicles based at a depot is used to complete a set of requests which consist of transporting items from a pickup location to a delivery location. The vehicles have multiple stacks operated using last-in-first-out (LIFO) loading which requires the vehicle to be rear-loaded and items can only be unloaded if they are closest to the back door. In the PDPTWMS-H, additional handling operations, referred to as rehandling, are allowed and an additional handling time might be incurred when rehandling items (by unloading and reloading items). The problem consists of determining the number of vehicles and the vehicle routes needed to complete the set of requests at minimal cost while respecting the possible handling operations. We model the PDPTWMS-H with a set-partitioning formulation and resort to branch-price-and-cut (BPC) for its solution. To solve the pricing problem, we derive a unified labeling algorithm that is able to cope with the different rehandling possibilities. The labeling algorithm keeps track about the information of on-board items such that symmetries with respect to both stacks and item positions are reduced. Extensive tests are performed on benchmark instances to assess the performance of the proposed BPC methodology and to provide insights on the impact of the rehandling flexibility on solution quality and time.

In this article, a dynamic-neighborhood-based switching PSO (DNSPSO) algorithm is proposed, where a new velocity updating mechanism is designed to adjust the personal best position and the global best position according to a distance-based dynamic neighborhood to make full use of the population evolution information among the entire swarm. In addition, a novel switching learning strategy is introduced to adaptively select the acceleration coefficients and update the velocity model according to the searching state at each iteration, thereby contributing to a thorough search of the problem space. Furthermore, the differential evolution algorithm is successfully hybridized with the particle swarm optimization (PSO) algorithm to alleviate premature convergence. A series of commonly used benchmark functions (including unimodal, multimodal, and rotated multimodal cases) is utilized to comprehensively evaluate the performance of the DNSPSO algorithm. The experimental results demonstrate that the developed DNSPSO algorithm outperforms a number of existing PSO algorithms in terms of the solution accuracy and convergence performance, especially for complicated multimodal optimization problems.

Algorithms inspired by nature became more popular in the last few years. They showed up powerful capability in solving optimization problems. This capability was obtained by their ability to be applied individually or by merging them with other algorithms or techniques. The Black Hole optimization algorithm is a nature-inspired algorithm that belongs to the meta-heuristic category. The Black-Hole algorithm (BH) simulates the black hole phenomenon which is formed from a star with massive size and very high gravitational power. The algorithm starts with a population of a specific size of possible solutions and then gets evaluated by selecting the best one as a black hole. In the suggested modifications, we have introduced a new idea for generating the stars absorbed by the black hole. The star movement towards the black hole has also been modified to increase exploration capabilities. The modified algorithm was used to prove its effectiveness in data clustering without any prior knowledge about the nature of the provided data. Several benchmark datasets and statistical techniques have been used to evaluate the performance of suggested modification. The experiment results promised that the improved algorithm can overcome popular optimization algorithms.

This article explores the problem of semisupervised affinity matrix learning, that is, learning an affinity matrix of data samples under the supervision of a small number of pairwise constraints (PCs). By observing that both the matrix encoding PCs, called pairwise constraint matrix (PCM) and the empirically constructed affinity matrix (EAM), express the similarity between samples, we assume that both of them are generated from a latent affinity matrix (LAM) that can depict the ideal pairwise relation between samples. Specifically, the PCM can be thought of as a partial observation of the LAM, while the EAM is a fully observed one but corrupted with noise/outliers. To this end, we innovatively cast the semisupervised affinity matrix learning as the recovery of the LAM guided by the PCM and EAM, which is technically formulated as a convex optimization problem. We also provide an efficient algorithm for solving the resulting model numerically. Extensive experiments on benchmark datasets demonstrate the significant superiority of our method over state-of-the-art ones when used for constrained clustering and dimensionality reduction. The code is publicly available at https://github.com/jyh-learning/LAM.

Cross-manifold clustering is a hard topic and many traditional clustering methods fail because of the cross-manifold structures. In this paper, we propose a Multiple Flat Projections Clustering (MFPC) to deal with cross-manifold clustering problems. In our MFPC, the given samples are projected into multiple subspaces to discover the global structures of the implicit manifolds. Thus, the cross-manifold clusters are distinguished from the various projections. Further, our MFPC is extended to nonlinear manifold clustering via kernel tricks to deal with more complex cross-manifold clustering. A series of non-convex matrix optimization problems in MFPC are solved by a proposed recursive algorithm. The synthetic tests show that our MFPC works on the cross-manifold structures well. Moreover, experimental results on the benchmark datasets show the excellent performance of our MFPC compared with some state-of-the-art clustering methods., Comment: 12 pages, 58 figures

The meaningful patterns embedded in high-dimensional multi-view data sets typically tend to have a much more compact representation that often lies close to a low-dimensional manifold. Identification of hidden structures in such data mainly depends on the proper modeling of the geometry of low-dimensional manifolds. In this regard, this article presents a manifold optimization-based integrative clustering algorithm for multi-view data. To identify consensus clusters, the algorithm constructs a joint graph Laplacian that contains denoised cluster information of the individual views. It optimizes a joint clustering objective while reducing the disagreement between the cluster structures conveyed by the joint and individual views. The optimization is performed alternatively over k -means and Stiefel manifolds. The Stiefel manifold helps to model the nonlinearities and differential clusters within the individual views, whereas k -means manifold tries to elucidate the best-fit joint cluster structure of the data. A gradient-based movement is performed separately on the manifold of each view so that individual nonlinearity is preserved while looking for shared cluster information. The convergence of the proposed algorithm is established over the manifold and asymptotic convergence bound is obtained to quantify theoretically how fast the sequence of iterates generated by the algorithm converges to an optimal solution. The integrative clustering on benchmark and multi-omics cancer data sets demonstrates that the proposed algorithm outperforms state-of-the-art multi-view clustering approaches.

Affective computing is a field of great interest in many computer vision applications, including video surveillance, behaviour analysis, and human-robot interaction. Most of the existing literature has addressed this field by analysing different sets of face features. However, in the last decade, several studies have shown how body movements can play a key role even in emotion recognition. The majority of these experiments on the body are performed by trained actors whose aim is to simulate emotional reactions. These unnatural expressions differ from the more challenging genuine emotions, thus invalidating the obtained results. In this paper, a solution for basic non-acted emotion recognition based on 3D skeleton and Deep Neural Networks (DNNs) is provided. The proposed work introduces three majors contributions. First, unlike the current state-of-the-art in non-acted body affect recognition, where only static or global body features are considered, in this work also temporal local movements performed by subjects in each frame are examined. Second, an original set of global and time-dependent features for body movement description is provided. Third, to the best of out knowledge, this is the first attempt to use deep learning methods for non-acted body affect recognition. Due to the novelty of the topic, only the UCLIC dataset is currently considered the benchmark for comparative tests. On the latter, the proposed method outperforms all the competitors.

In this paper, we address Emotion Recognition in Conversation (ERC) where conversational data are presented in a multimodal setting. Psychological evidence shows that self and inter-speaker influence are two central factors to emotion dynamics in conversation. State-of-the-art models do not effectively synthesise these two factors. Therefore, we propose an Adapted Dynamic Memory Network (A-DMN) where self and inter-speaker influences are modelled individually and further synthesised oriented towards the current utterance. Specifically, we model the dependency of the constituent utterances in a dialogue video using a global RNN to capture inter-speaker influence. Likewise, each speaker is assigned an RNN to capture their self influence. Afterwards, an Episodic Memory Module is devised to extract contexts for self and inter-speaker influence and synthesise them to update the memory. This process repeats itself for multiple passes until a refined representation is obtained and used for final prediction. Additionally, we explore cross-modal fusion in the context of multimodal ERC, and propose a convolution-based method which proves effective in extracting local interactions and computationally efficient. Extensive experiments demonstrate that A-DMN outperforms the state-of-the-art models on benchmark datasets.

Attributed networks are ubiquitous in the real world, such as social networks. Therefore, many researchers take the node attributes into consideration in the network representation learning to improve the downstream task performance. In this article, we mainly focus on an untouched ``oversmoothing'' problem in the research of the attributed network representation learning. Although the Laplacian smoothing has been applied by the state-of-the-art works to learn a more robust node representation, these works cannot adapt to the topological characteristics of different networks, thereby causing the new oversmoothing problem and reducing the performance on some networks. In contrast, we adopt a smoothing parameter that is evaluated from the topological characteristics of a specified network, such as small worldness or node convergency and, thus, can smooth the nodes' attribute and structure information adaptively and derive both robust and distinguishable node features for different networks. Moreover, we develop an integrated autoencoder to learn the node representation by reconstructing the combination of the smoothed structure and attribute information. By observation of extensive experiments, our approach can preserve the intrinsical information of networks more effectively than the state-of-the-art works on a number of benchmark datasets with very different topological characteristics.

Recently, deep learning models have been successfully employed in video-based affective computing applications. One key application is automatic depression recognition from facial expressions. State-of-the-art approaches to recognize depression typically explore spatial and temporal information individually, by using convolutional neural networks (CNNs) to analyze appearance information, and then by either mapping feature variations or averaging the depression level over video frames. This approach has limitations to represent dynamic information that can help to discriminate between depression levels. In contrast, 3D CNN-based models can directly encode the spatio-temporal relationships, although these models rely on fixed-range temporal information and single receptive field. This approach limits the ability to capture facial expression variations with diverse ranges, and the exploitation of diverse facial areas. In this paper, a novel 3D CNN architecture the Multiscale Spatiotemporal Network (MSN) is introduced to effectively represent facial information related to depressive behaviours. The basic structure of the model is composed of parallel convolutional layers with different temporal depths and sizes of receptive field, which allows the MSN to explore a wide range of spatio-temporal variations in facial expressions. Experimental results on two benchmark datasets show that our MSN is effective, outperforming state-of-the-art methods in automatic depression recognition.

Detection of Alzheimer's Disease (AD) from neuroimaging data such as MRI through machine learning has been a subject of intense research in recent years. Recent success of deep learning in computer vision has progressed such research further. However, common limitations with such algorithms are reliance on a large number of training images, and requirement of careful optimization of the architecture of deep networks. In this paper, we attempt solving these issues with transfer learning, where the state-of-the-art VGG architecture is initialized with pre-trained weights from large benchmark datasets consisting of natural images. The network is then fine-tuned with layer-wise tuning, where only a pre-defined group of layers are trained on MRI images. To shrink the training data size, we employ image entropy to select the most informative slices. Through experimentation on the ADNI dataset, we show that with training size of 10 to 20 times smaller than the other contemporary methods, we reach state-of-the-art performance in AD vs. NC, AD vs. MCI, and MCI vs. NC classification problems, with a 4% and a 7% increase in accuracy over the state-of-the-art for AD vs. MCI and MCI vs. NC, respectively. We also provide detailed analysis of the effect of the intelligent training data selection method, changing the training size, and changing the number of layers to be fine-tuned. Finally, we provide Class Activation Maps (CAM) that demonstrate how the proposed model focuses on discriminative image regions that are neuropathologically relevant, and can help the healthcare practitioner in interpreting the model's decision making process.

The intrinsic permeability is a crucial parameter to characterise and quantify fluid flow through porous media. However, this parameter is typically uncertain, even if the geometry of the pore structure is available. In this paper, we perform a comparative study of experimental, semi-analytical and numerical methods to calculate the permeability of a regular porous structure. In particular, we use the Kozeny-Carman relation, different homogenisation approaches (3D, 2D, very thin porous media and pseudo 2D/3D), pore-scale simulations (lattice Boltzmann method, Smoothed Particle Hydrodynamics and finite-element method) and pore-scale experiments (microfluidics). A conceptual design of a periodic porous structure with regularly positioned solid cylinders is set up as a benchmark problem and treated with all considered methods. The results are discussed with regard to the individual strengths and limitations of the used methods. The applicable homogenisation approaches as well as all considered pore-scale models prove their ability to predict the permeability of the benchmark problem. The underestimation obtained by the microfluidic experiments is analysed in detail using the lattice Boltzmann method, which makes it possible to quantify the influence of experimental setup restrictions., Deutsche Forschungsgemeinschaft, Projekt DEAL

Full projector compensation aims to modify a projector input image to compensate for both geometric and photometric disturbance of the projection surface. Traditional methods usually solve the two parts separately and may suffer from suboptimal solutions. In this paper, we propose the first end-to-end differentiable solution, named CompenNeSt++, to solve the two problems jointly. First, we propose a novel geometric correction subnet, named WarpingNet, which is designed with a cascaded coarse-to-fine structure to learn the sampling grid directly from sampling images. Second, we propose a novel photometric compensation subnet, named CompenNeSt, which is designed with a siamese architecture to capture the photometric interactions between the projection surface and the projected images, and to use such information to compensate the geometrically corrected images. By concatenating WarpingNet with CompenNeSt, CompenNeSt++ accomplishes full projector compensation and is end-to-end trainable. Third, to improve practicability, we propose a novel synthetic data-based pre-training strategy to significantly reduce the number of training images and training time. Moreover, we construct the first setup-independent full compensation benchmark to facilitate future studies. In thorough experiments, our method shows clear advantages over prior art with promising compensation quality and meanwhile being practically convenient., Source code: https://github.com/BingyaoHuang/CompenNeSt-plusplus. arXiv admin note: text overlap with arXiv:1908.06246, arXiv:1904.04335

Training an Artificial Neural Network (ANN) algorithm is not trivial, which requires optimizing a set of weights and biases that increase dramatically with the increasing capacity of the neural network resulting in such hard optimization problems. Essentially, over recent decades, stochastic search algorithms have shown remarkable abilities for addressing hard optimization problems. On the other hand, pragmatically, abundant real-world problems suffer from the imbalance problem, where the distribution of data varies considerably among classes resulting in more training biases and variances which degrades the performance of the learning algorithm. This paper introduces three stochastic and metaheuristic algorithms for training the Multilayer Perceptron (MLP) neural network to solve the problem of imbalanced classifications. The utilized algorithms are the Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and the Salp Swarm Algorithm (SSA). The proposed GWO-MLP, PSO-MLP, and SSA-MLP are trained based on different objective functions; accuracy, f1-score, and g-mean. Whereas, it is evaluated based on 10 benchmark imbalanced datasets. The results show an advantage for f1-score, and g-mean fitness functions over the accuracy when the datasets are imbalanced.

International audience; Existing studies in black-box optimization for machine learning suffer from low generalizability, caused by a typically selective choice of problem instances used for training and testing different optimization algorithms. Among other issues, this practice promotes overfitting and poor-performing user guidelines. To address this shortcoming, we propose in this work a benchmark suite, OptimSuite, which covers a broad range of black-box optimization problems, ranging from academic benchmarks to real-world applications, from discrete over numerical to mixed-integer problems, from small to very large-scale problems, from noisy over dynamic to static problems, etc. We demonstrate the advantages of such a broad collection by deriving from it Automated Black Box Optimizer (ABBO), a general-purpose algorithm selection wizard. Using three different types of algorithm selection techniques, ABBO achieves competitive performance on all benchmark suites. It significantly outperforms previous state of the art on some of them, including YABBOB and LSGO. ABBO relies on many high-quality base components. Its excellent performance is obtained without any task-specific parametrization. The OptimSuite benchmark collection, the ABBO wizard and its base solvers have all been merged into the open-source Nevergrad platform, where they are available for reproducible research.

In recent years, Jaya optimization algorithm has been successfully applied in several optimization problems. This paper presents a novel feature selection (FS) approach based on Jaya optimization algorithm (FSJaya) along with supervised machine learning techniques to select the optimal features. This approach uses a search technique to find the best suitable features by updating the worst features to reduce the dimensions of the feature space. This improves the performance of supervised machine learning techniques. The effectiveness of the proposed approach is evaluated for ten benchmark datasets and compared with several FS approaches such as FS using genetic algorithm (FSGA), FS using particle swarm optimization algorithm (FSPSO), and FS using differential evolutionary (FSDE). The experimental result has shown that the average classification accuracy of FSJaya on most of the datasets is superior over the existing methods such as FSGA, FSPSO, and FSDE. The proof of statistical significance of the proposed approach has been validated by using Friedman and Holm test. This proposed approach is found efficient in selecting an optimal subset of features as compared to other counterparts.

Sensory data are critical for soft robot perception. However, integrating sensors to soft robots remains challenging due to their inherent softness. An alternative approach is indirect sensing through an estimation scheme, which uses robot dynamics and available measurements to estimate variables that would have been measured by sensors. Nevertheless, developing an adequately effective estimation scheme for soft robots is not straightforward. First, it requires a mathematical model; modeling of soft robots is analytically demanding due to their complex dynamics. Second, it should perform multimodal sensing for both internal and external variables, with minimal sensors, and finally, it must be robust against sensor faults. In this article, we propose a recurrent neural network-based adaptive unscented Kalman filter (RNN-AUKF) architecture to estimate the proprioceptive state and exteroceptive unknown input of a pneumatic-based soft finger. To address the challenge in modeling soft robots, we adopt a data-driven approach using RNNs. Then, we interconnect the AUKF with an unknown input estimator to perform multimodal sensing using a single embedded flex sensor. We also prove mathematically that the estimation error is bounded with respect to sensor degradation (noise and drift). Experimental results show that the RNN-AUKF achieves a better overall performance in terms of accuracy and robustness against the benchmark method. The proposed scheme is also extended to a multifinger soft gripper and is robust against out-of-distribution sensor dynamics. The outcomes of this research have immense potentials in realizing a robust multimodal indirect sensing in soft robots.

In this paper, we propose a general framework termed "Centroid Estimation with Guaranteed Efficiency" (CEGE) for Weakly Supervised Learning (WSL) with incomplete, inexact, and inaccurate supervision. The core of our framework is to devise an unbiased and statistically efficient risk estimator that is applicable to various weak supervision. Specifically, by decomposing the loss function (e.g., the squared loss and hinge loss) into a label-independent term and a label-dependent term, we discover that only the latter is influenced by the weak supervision and is related to the centroid of the entire dataset. Therefore, by constructing two auxiliary pseudo-labeled datasets with synthesized labels, we derive unbiased estimates of centroid based on the two auxiliary datasets, respectively. These two estimates are further linearly combined with a properly decided coefficient which makes the final combined estimate not only unbiased but also statistically efficient. This is better than some existing methods that only care about the unbiasedness of estimation but ignore the statistical efficiency. The good statistical efficiency of the derived estimator is guaranteed as we theoretically prove that it acquires the minimum variance when estimating the centroid. As a result, intensive experimental results on a large number of benchmark datasets demonstrate that our CEGE generally obtains better performance than the existing approaches related to typical WSL problems including semi-supervised learning, positive-unlabeled learning, multiple instance learning, and label noise learning.

As an essential problem in computer vision, salient object detection (SOD) has attracted an increasing amount of research attention over the years. Recent advances in SOD are predominantly led by deep learning-based solutions (named deep SOD). To enable in-depth understanding of deep SOD, in this paper, we provide a comprehensive survey covering various aspects, ranging from algorithm taxonomy to unsolved issues. In particular, we first review deep SOD algorithms from different perspectives, including network architecture, level of supervision, learning paradigm, and object-/instance-level detection. Following that, we summarize and analyze existing SOD datasets and evaluation metrics. Then, we benchmark a large group of representative SOD models, and provide detailed analyses of the comparison results. Moreover, we study the performance of SOD algorithms under different attribute settings, which has not been thoroughly explored previously, by constructing a novel SOD dataset with rich attribute annotations covering various salient object types, challenging factors, and scene categories. We further analyze, for the first time in the field, the robustness of SOD models to random input perturbations and adversarial attacks. We also look into the generalization and difficulty of existing SOD datasets. Finally, we discuss several open issues of SOD and outline future research directions., Published on IEEE TPAMI. All the saliency prediction maps, our constructed dataset with annotations, and codes for evaluation are publicly available at \url{https://github.com/wenguanwang/SODsurvey}

Multivariate time series (MTSs) are widely found in many important application fields, for example, medicine, multimedia, manufacturing, action recognition, and speech recognition. The accurate classification of MTS has become an important research topic. Traditional MTS classification methods do not explicitly model the temporal difference information of time series, which is, in fact, important and reflects the dynamic evolution information. In this article, the difference-guided representation learning network (DGRL-Net) is proposed to guide the representation learning of time series by dynamic evolution information. The DGRL-Net consists of a difference-guided layer and a multiscale convolutional layer. First, in the difference-guided layer, we propose a difference gating LSTM to model the time dependency and dynamic evolution of the time series to obtain feature representations of both raw and difference series. Then, these two representations are used as two input channels of the multiscale convolutional layer to extract multiscale information. Extensive experiments demonstrate that the proposed model outperforms state-of-the-art methods on 18 MTS benchmark datasets and achieves competitive results on two skeleton-based action recognition datasets. Furthermore, the ablation study and visualized analysis are designed to verify the effectiveness of the proposed model.

The covariance matrix self-adaptation evolution strategy with repelling subpopulations (RS-CMSA-ES) is one of the most successful multimodal optimization methods currently available. However, some of its components may become inefficient in certain situations. This study introduces the second variant of this method, called RS-CMSA-ESII. It improves the adaptation schemes for the normalized taboo distances of the archived solutions and the covariance matrix of the subpopulation, the termination criteria for the subpopulations, and the way in which the infeasible solutions are treated. It also improves the time complexity of RS-CMSA-ES by updating the initialization procedure of a subpopulation and developing a more accurate metric for determining critical taboo regions. The effects of these modifications are illustrated by designing controlled numerical simulations. RS-CMSA-ESII is then compared with the most successful and recent niching methods for multimodal optimization on a widely adopted test suite. The results obtained reveal the superiority of RS-CMSA-ESII over these methods, including the winners of the competition on niching methods for multimodal optimization in previous years. Besides, this study extends RS-CMSA-ESII to dynamic multimodal optimization and compares it with a few recently proposed methods on the modified moving peak benchmark functions.

Existing face anti-spoofing models using deep learning for multi-modality data suffer from low generalization in the case of using variety of presentation attacks such as 2D printing and high-precision 3D face masks. One of the main reasons is that the non-linearity of multi-spectral information used to preserve the intrinsic attributes between a real and a fake face are not well extracted. To address this issue, we propose a multi-modility data based two-stage cascade framework for face anti-spoofing. The proposed framework has two advantages. Firstly, we design a two-stage cascade architecture that can selectively fuse low-level and high-level features from different modalities to improve feature representation. Secondly, we use multi-modality data to construct a distance-free spectral on RGB and infrared (IR) to augment the non-linearity of data. The presented data fusion strategy is different from popular fusion approaches, since it can strengthen discrimination ability of network models on physical attribute features than identity structure features under certain constraints. In addition, a multi-scale patch based weighted fine-tuning strategy is designed to learn each specific local face region. Experimental results show that the proposed framework achieves better performance than other state-of-the-art methods on both benchmark datasets and self-established datasets, especially on multi-material masks spoofing.

The driving force behind content dissemination in Social network (SN) is the users’ interest in the content, which is strongly reflected in their interactions. Obviously, user interest varies with the disseminated content. Consequently, the dynamic interest results in decomposing SN into dynamic user clusters “interest groups”. The objective of this work is to rank interest-based communities using influence propagation. The contribution of this work is threefold: First, to highlight the significance of the indirect influence among interest-based user groups. Second, to study its impact on content dissemination capability. Third, to propose an ultimate ranking model (UltRank) that uniquely considers direct and indirect influences which are reflected in a new reachability metric that considers: 1. Distance among interest groups. 2. Percentage of reachable interest groups. 3. Percentage of reachable nodes. UltRank model has been evaluated in comprehensive experiments. First, clustering quality perspective, the Silhouette coefficient for the identified interest groups is on average 0.996 and the Jaccard coefficient of 97% of different interest groups members equals 0. Second, ranking capability perspective, UltRank model can rank up to 91% of interest groups in SN. Finally, ranking effectiveness perspective, UltRank ranking list has a competing network coverage results against the other benchmark approaches.

With the increasing demand of users for personalized social services, social recommendation (SR) has been an important concern in academia. However, current research on SR universally faces two main challenges. On the one hand, SR lacks the considerable ability of robust online data management. On the other hand, SR fails to take the ambiguity of preference feedback into consideration. To bridge these gaps, a deep learning-embedded social Internet of Things (IoT) is proposed for ambiguity-aware SR (SIoT-SR). Specifically, a social IoT architecture is developed for social computing scenarios to guarantee reliable data management. A deep learning-based graph neural network model that can be embedded into the model is proposed as the core algorithm to perform ambiguity-aware SR. This design not only provides proper online data sensing and management but also overcomes the preference ambiguity problem in SR. To evaluate the performance of the proposed SIoT-SR, two real-world datasets are selected to establish experimental scenarios. The method is assessed using three different metrics, selecting five typical methods as benchmarks. The experimental results show that the proposed SIoT-SR performs better than the benchmark methods by at least 10% and has good robustness.

Classification is used to categorize data and produce decisions for several domains. To improve the accuracy of classification, researchers have tended to hybridize the neural network with other metaheuristic algorithms in order to better exploit and explore the search space and thereby solve many different classification problems in an effective manner. The hybridization of algorithms is now commonplace and has resulted in the creation of novel methods that are more effective in comparison with those that employ a sole algorithm. Therefore, in this paper, a hybridization approach is employed to utilize the African buffalo optimization (ABO) algorithm as an optimizer to adjust the weights of the probabilistic neural network (PNN). The effectiveness of the proposed (ABO-PNN) method is investigated by applying it to several different classification problems. The efficiency of the ABO algorithm is assessed based on the PNN training results produced and its performance is compared with that of different types of optimization algorithm. The performance of the proposed algorithm in terms of classification accuracy is tested on 11 benchmark datasets. The results show that the ABO is better than the firefly algorithm (FA) in terms of both classification accuracy and convergence speed.

In this work, the problem of ensuring profitability for multiple sensor-owners in sensor-cloud, while satisfying the service-requirements of end-users, is studied. In traditional sensor-cloud, Sensor-Cloud-Service-Provider (SCSP) solely dictates the service-provisioning process. However, the SCSP cannot always ensure high profits for sensor-owners, who incur significant maintenance costs for their sensor-nodes. Contrarily, it is highly essential to meet the Quality-of-Information (QoI) requirements of end-users to ensure their service-satisfaction. In the existing literature, researchers proposed few node-allocation schemes which neither consider the cost incurred by sensor-owners nor the QoI of sensed-data in sensor-cloud. To address this problem, a strategic resource-allocation scheme, named RACE, is proposed, which introduces the participation of sensor-owners in the node-allocation process. Firstly, utility theory is used to calculate the optimum number-of-nodes to be allocated for a service. Thereafter, single-leader-multiple-followers Stackelberg game is formulated to decide the number-of-nodes to be contributed by each sensor-owner and the price to be charged. Through simulations, we observed that, using RACE, the profits of the sensor-owners and those of the SCSP increase by 86.11-89.26% and 41.95-80.82%, respectively, as compared to existing benchmark schemes, while considering that each sensor-node is capable of serving multiple applications simultaneously. Moreover, service-availability in sensor-cloud increases by 31.70-96.96% using RACE.

Visual tracking is typically solved as a discriminative learning problem that usually requires high-quality samples for online model adaptation. It is a critical and challenging problem to evaluate the training samples collected from previous predictions and employ sample selection by their quality to train the model. To tackle the above problem, we propose a joint discriminative learning scheme with the progressive multistage optimization policy of sample selection for robust visual tracking. The proposed scheme presents a novel time-weighted and detection-guided self-paced learning strategy for easy-to-hard sample selection, which is capable of tolerating relatively large intraclass variations while maintaining interclass separability. Such a self-paced learning strategy is jointly optimized in conjunction with the discriminative tracking process, resulting in robust tracking results. Experiments on the benchmark datasets demonstrate the effectiveness of the proposed learning framework.

Annotated corpus can greatly assist in the natural language processing field. For example, computers can understand more of the document context, and indexing and clustering in information retrieval can be done precisely with less or no ambiguity of words. However, there are only a few annotated corpora in Malay language, which are not publicly shared. In this paper, we delve into analysing and annotating Malay translated hadith documents in terms of tagging and entities. There are three phases, which are manual filtering and cleaning, analysing the corpus and creating the benchmark. As the result, an analysis and benchmark of Malay translated hadith corpus were produced in term of part-of-speech and named entities tags that follows the Zipf’s law distribution.

Solving constrained multiobjective optimization problems brings great challenges to an evolutionary algorithm, since it simultaneously requires the optimization among several conflicting objective functions and the satisfaction of various constraints. Hence, how to adjust the tradeoff between objective functions and constraints is crucial. In this article, we propose a dynamic selection preference-assisted constrained multiobjective differential evolutionary (DE) algorithm. In our approach, the selection preference of each individual is suitably switching from the objective functions to constraints as the evolutionary process. To be specific, the information of objective function, without considering any constraints, is extracted based on Pareto dominance to maintain the convergence and diversity by exploring the feasible and infeasible regions; while the information of constraint is used based on constrained dominance principle to promote the feasibility. Then, the tradeoff in these two kinds of information is adjusted dynamically, by emphasizing the utilization of objective functions at the early stage and focusing on constraints at the latter stage. Furthermore, to generate the promising offspring, two DE operators with distinct characteristics are selected as components of the search algorithm. Experiments on four test suites including 56 benchmark problems indicate that the proposed method exhibits superior or at least competitive performance, in comparison with other well-established methods.

The objective of the design and operation of any water distribution network (WDN) includes meeting the desired demand at sufficient pressure at all nodes. However, this requires situational awareness; in other words, the knowledge of system state variables such as pressure and flow throughout the network. In this work, a hybrid approach is developed for sensor placement (SP) and state estimation (SE) that exploits the underlying correlation structure in the data, along with the principles governing the flow through circular pipes. The problem of SP in WDN is addressed since measuring the state variables throughout the network is not practical. The problem of SE that maps to a matrix completion problem under certain physical and logical constraints is solved later. The completed matrix represents the state of WDN at any given time. Benchmark networks used in literature were used to evaluate the proposed approach. The mean absolute percentage error (MAPE) of less than 5% was obtained while estimating the head available at nodes. The knowledge of the states in the entire network could help operate the network adaptively.

Feature selection is of great importance to make prediction for process variables in industrial production. An embedded feature selection method, based on relevance vector machines with an approximated marginal likelihood function, is proposed in this study. By setting hierarchical prior distributions over the model weights and the parameters of the automatic relevance determination kernel function, respectively, a Gaussian approximation method is designed to approximate the intractable exact marginal likelihood by using the law of the total expectation and the total covariance. Furthermore, in this study, the joint posterior distribution over the model weights and the kernel parameters is estimated by combining the Gibbs sampling with a Laplace approximation. Thus, feature selection is performed by examining the posterior over the kernel parameters. To verify the performance of the proposed method, a series of benchmark datasets and two practical industrial datasets are employed. The experimental results demonstrate that the proposed method not only produces higher prediction accuracy than other methods but also performs better in feature selection, especially in industrial case.

With time-varying workloads and service requests, cloud-based software services necessitate adaptive resource allocation for guaranteeing Quality-of-Service (QoS) and reducing resource costs. However, due to the ever-changing system states, resource allocation for cloud-based software services faces huge challenges in dynamics and complexity. The traditional approaches mostly rely on expert knowledge or numerous iterations, which might lead to weak adaptiveness and extra costs. Moreover, existing RL-based methods target the environment with the fixed workload, and thus they are unable to effectively fit in the real-world scenarios with variable workloads. To address these important challenges, we propose a Prediction-enabled feedback Control with Reinforcement learning based resource Allocation (PCRA) method. First, a novel Q-value prediction model is designed to predict the values of management operations (by Q-values) at different system states. The model uses multiple prediction learners for making accurate Q-value prediction by integrating the Q-learning algorithm. Next, the objective resource allocation plans can be found by using a new feedback-control based decision-making algorithm. Using the RUBiS benchmark, simulation results demonstrate that the PCRA chooses the management operations of resource allocation with 93.7% correctness. Moreover, the PCRA achieves optimal/near-optimal performance, and it outperforms the classic ML-based and rule-based approaches by 5~7% and 10~13%, respectively.

We present a new branch-and-bound algorithm for the Quadratic Multiple Knapsack Problem. A key component of our algorithm is a new upper bound that divides the pairwise item values among individual items, estimates the maximum potential value contributed by each individual item, and calculates the upper bound via a transportation model. A local search is used to adjust the division of pairwise item values in order to improve the upper bound. Reduced costs from the solution of the transportation problem are used to forbid some item-to-knapsack assignments. Information from the upper bound is also used in selecting the item to branch on next. Computational experiments are carried out on three sets of benchmark instances from the literature, two sets of smaller instances with 20–35 items and 3–10 knapsacks per instance, and one set of larger instances with 40–60 items and 3–10 knapsacks per instance. Our algorithm finds and verifies optimal solutions for all small benchmark instances in less than one hour of CPU time apiece and outperforms the best previously proposed algorithms for the problem in terms of both the average computation time and the number of instances for which optimality is verified. For the larger benchmark instances, our algorithm obtains smaller average gaps than the best previously proposed methods.

Safety is one of the important topics in the field of civil aviation. Auxiliary Power Unit (APU) is one of important components in aircraft, which provides electrical power and compressed air for aircraft. The hazards in APU are prone to cause economic losses and even casualties. So, actively identifying the hazards in APU before an accident occurs is necessary. In this paper, a Hybrid Deep Neural Network (HDNN) based on multi-time window convolutional neural network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) neural network is proposed for active hazard identification of APU in civil aircraft. In order to identify the risks caused by different types of failures, the proposed HDNN simultaneously integrates three CNN-BiLSTM basic models with different time window sizes in parallel by using a fully connected neural network. The CNN-BiLSTM basic model can automatically extract features representing the system state from the input data and learn the time information of irregular trends in the time series data. Nine benchmark models are compared with the proposed HDNN. The comparison results show that the proposed HDNN has the highest identification accuracy. The HDNN has the most stable identification performance for data with imbalanced samples.

This paper introduces two novel algorithms for thread migrations, named CIMAR (Core-aware Interchange and Migration Algorithm with performance Record –IMAR–) and NIMAR (Node-aware IMAR), and a new algorithm for the migration of memory pages, LMMA (Latency-based Memory pages Migration Algorithm), in the context of Non-Uniform Memory Access (NUMA) systems. This kind of system has complex memory hierarchies that present a challenging problem in extracting the best possible performance, where thread and memory mapping play a critical role. The presented algorithms gather and process the information provided by hardware counters to make decisions about the migrations to be performed, trying to find the optimal mapping. They have been implemented as a user space tool that looks for improving the system performance, particularly in, but not restricted to, scenarios where multiple programs with different characteristics are running. This approach has the advantage of not requiring any modification on the target programs or the Linux kernel while keeping a low overhead. Two different benchmark suites have been used to validate our algorithms: The NAS parallel benchmark, mainly devoted to computational routines, and the LevelDB database benchmark focused on read–write operations. These benchmarks allow us to illustrate the influence of our proposal in these two important types of codes. Note that those codes are state-of-the-art implementations of the routines, so few improvements could be initially expected. Experiments have been designed and conducted to emulate three different scenarios: a single program running in the system with full resources, an interactive server where multiple programs run concurrently varying the availability of resources, and a queue of tasks where granted resources are limited. The proposed algorithms have been able to produce significant benefits, especially in systems with higher latency penalties for remote accesses. When more than one benchmark is executed simultaneously, performance improvements have been obtained, reducing execution times up to 60%. In this kind of situation, the behaviour of the system is more critical, and the NUMA topology plays a more relevant role. Even in the worst case, when isolated benchmarks are executed using the whole system, that is, just one task at a time, the performance is not degraded This research work has received financial support from the Ministerio de Ciencia e Innovación, Spain within the project PID2019-104834GB-I00. It was also funded by the Consellería de Cultura, Educación e Ordenación Universitaria of Xunta de Galicia (accr. 2019–2022, ED431G 2019/04 and reference competitive group 2019–2021, ED431C 2018/19) SI

For the path coverage testing of a Message-Passing Interface (MPI) program, test data generation based on an evolutionary optimization algorithm (EOA) has been widely known. However, during the use of the above technique, it is necessary to evaluate the fitness of each evolutionary individual by executing the program, which is generally computationally expensive. In order to reduce the computational cost, this paper proposes a method of integrating an ensemble surrogate model's estimation into the process of generating test data. The proposed method first produces a number of test inputs using an EOA, and forms a training set together with their real fitness. Then, this paper trains an ensemble surrogate model (ESM) based on the training set, which is employed to estimate the fitness of each individual. Finally, a small number of individuals with good estimations are selected to further execute the program, so as to have their real fitness for the subsequent evolution. This paper applies the proposed method to seven benchmark MPI programs, which is compared with several state-of-the-art approaches. The experimental results show that the proposed method can generate test data with significantly low computational cost.

The redundancy allocation problem (RAP) is one of the most commonly used approaches for improving the reliability of systems. The present study investigates the most recent variant of RAP, namely, the RAP with heterogeneous components under the mixed redundancy strategy. An exact branch-and-price (BP) algorithm is proposed for the problem that solves in less than one CPU second all the benchmark instances reported in the literature. The algorithm is not only the first exact one proposed for the RAP with a mixed strategy but also the first application of the BP algorithm in the literature of reliability optimization problems. The proposed BP algorithm is also capable of solving other variants of RAP, including RAPs with either an active or a standby strategy, making it possible to investigate the advantages of the mixed strategy over the active and standby strategies. Finally, a new set of instances is generated to evaluate the performance of the proposed algorithm in dealing with more challenging instances.