Search

In this paper, we propose an unsupervised textural-intensity-based natural image quality evaluator (TI-NIQE) by modelling the texture, structure and naturalness of an image. In detail, an effective quality-aware feature named as textural intensity (TI) is proposed in this paper to detect image texture. The image structure is captured by the distribution of gradients and basis images. The naturalness is characterized through the distributions of the locally mean subtracted and contrast normalized (MSCN) coefficients and the products of pairs of the adjacent MSCN coefficients. Furthermore, a new application pattern of image quality assessment (IQA) measures is proposed by taking the quality scores as the essential input of the recognition model. Using statistics of video quality scores computed by TI-NIQE as input features, an automatic IQA-based visual recognition model is proposed for the condition recognition in rotary kiln. Extensive experiments on benchmark datasets demonstrate that TI-NIQE shows better performance both in accuracy and computational complexity than other state-of-the-art unsupervised IQA methods, and experimental results on real-world data show that the recognition model has high prediction accuracy for condition recognition in rotary kiln.

The reliability of the insulated gate bipolar transistors (IGBTs) is essential to the stable operation of the modular multilevel converter (MMC) system. However, there are a large number of IGBTs in the MMC system and the open-circuit faults of IGBTs are usually so hidden that it is difficult to find. Therefore, this article proposes a fault diagnosis framework based on temporal convolutional network (TCN) integrating adaptive chirp mode decomposition (ACMD) and silhouette coefficient (SC). First, ACMD is used to extract and reconstruct signal components from the original signal. Then, in order to avoid artificial selection of signal components, silhouette coefficient is introduced to characterize the importance of each component. Finally, the TCN model automatically extracts the features of the signal components and outputs the classification results. The main contributions are as follows: (1) A complete fault diagnosis framework that can adaptively extract features and perform fault classification is proposed in the paper. (2) For the MMC using the carrier-phase-shifted pulsewidth modulation strategy, the fault can be located to the IGBT by the output current. (3) Under certain noise conditions, the fault diagnosis proposed in the paper method still has good robustness. (4) The signal visualization of different residual blocks and channels explains the working mechanism of the AMCD-SC-TCN framework.

The construction of smart city makes information interconnection play an increasingly important role in intelligent surveillance systems. Especially the interconnection among massive cameras is the key to realizing the evolution from current fragmented monitoring to interconnection surveillance. However, it remains a challenging problem in practical systems due to large sensor quantity, various camera types, and complex spatial layout. Aimed at this problem, this paper proposes a novel multi-camera joint spatial self-organization approach, which realizes interconnection surveillance by unifying cameras into one imaging space. Differing from existing back-end data association strategy, our method takes front-end data calibration as a breakthrough to relate surveillance data. Specifically, this paper first initials camera spatial parameter by sequence complementary feature integration. Through integrating complementarity and redundancy among sequence features, our method has robustness under scene dynamic changes and noise. Then, we propose a multi-camera joint optimization method based on common monitoring coverage correlation analysis to estimate a more accurate relative relationship. By leveraging the two strategies, the spatial relationship and visual data association across monitoring cameras are returned finally. Our system organizes all cameras into a unified imaging space by itself. Extensive experimental evaluations on an actual campus environment demonstrate our method achieves remarkable performance.

This paper investigates the task assignment and guidance issues of unmanned surface vessels (USVs) interception. When the USVs formation is invaded by some moving objects during its escort, it is necessary for the unmanned systems to assign defenders to prevent attackers approaching the vulnerable target in antagonistic scenarios. This action requires efficient guidance and task assignment strategies. With this in mind, this paper presents the Integral Proportional Navigation Guidance (IPNG) with Tabu Dynamic Consensus-Based Auction Algorithm (TDCBAA) in marine interception scenario. First, IPNG is introduced in the interception game considering the USV kinematic model, which can effectively reduce the individual interception time. Second, a new bidding function is designed for moving objects interception with the consideration of the attackers’ types, positions and interception time. Finally, a TDCBAA is designed to solve the task assignment subproblem, resulting in a shorter overall interception time and a higher interception success rate. Simulations demonstrate that the proposed algorithm can optimize the allocation of defenders in real-time and intercept the attackers more quickly compared with other classical algorithms, which is more suitable in situations where attackers are approaching from all directions.

As the impact of crime on the lives of residents has increased, there are a number of methods for predicting where crime will occur. They tend to explore only the association established between a single factor and the distribution of crime. In order to more accurately and quickly visualize and predict crime distribution in different neighborhoods, and to provide a basis for security planning and design by planning designers, this paper uses GAN neural networks to build a prediction model of city floor plans and corresponding crime distribution maps. We take Philadelphia as the research sample, use more than 2 million crime information of Philadelphia from 2006 to 2018 to draw the crime hotspot distribution map, and collect the corresponding map of Philadelphia, and train the model for predicting the crime rate of the city with more than two thousand sets of one-to-one corresponding images as the training set. When the training is complete, a floor plan can be fed directly to the model, and the model will immediately feed back a hotspot map reflecting the crime distribution. Using the untrained Philadelphia data as the test set, the model can accurately predict crime concentration areas and the predicted crime concentration areas are similar to the concentration areas considered in previous studies. With the feedback from the model, the city layout can be adjusted and the crime rate can be greatly reduced when the simulated city planner tunes into the city plan. In addition the ideas in this paper can be applied as a set of methodologies to predict other relevant urban characteristic parameters and visualize them.

In this paper, an efficient artificial neural group synchronization is proposed for secured neural key exchange over public channels. To share the key over a public network, two Artificial Neural Networks (ANNs) are coordinated by mutual learning. The primary issue of neural coordination is assessing the synchronization of two parties’ ANNs in the absence of weights from the other. There is a delay in coordination measurement in existing techniques, which affects the confidentiality of neural coordination. Furthermore, research into the mutual learning of a cluster of ANNs is limited. This paper introduces a mutual learning methodology for measuring the entire synchronization of the set of ANNs quickly and efficiently. The measure of coordination is determined by the frequency with which the two networks have had the same outcome in prior rounds. When a particular threshold is reached, the hash is used to decide whether all networks are properly coordinated. The modified methodology uses has value of the weight vectors to achieve full coordination between two communicating entities. This technique has several advantages, including (1) Generation of session key via complete binary tree-based group mutual neural synchronization of ANNs over the public channel. (2) Unlike existing methods, the suggested method allows two communication entities to recognize full coordination faster. (3) Brute force, geometric, impersonation, and majority attacks are all considered in this proposed scheme. Tests to validate the performance of the proposed methodology are carried out, and the results show that the proposed methodology outperforms similar approaches already in use.

This paper describes a novel DNN-based system, named PD3net , that detects multiple people from a single depth image, in real time. The proposed neural network processes a depth image and outputs a likelihood map in image coordinates, where each detection corresponds to a Gaussian-shaped local distribution, centered at each person’s head. This likelihood map encodes both the number of detected people as well as their position in the image, from which the 3D position can be computed. The proposed DNN includes spatially separated convolutions to increase performance, and runs in real-time with low budget GPUs. We use synthetic data for initially training the network, followed by fine tuning with a small amount of real data. This allows adapting the network to different scenarios without needing large and manually labeled image datasets. Due to that, the people detection system presented in this paper has numerous potential applications in different fields, such as capacity control, automatic video-surveillance, people or groups behavior analysis, healthcare or monitoring and assistance of elderly people in ambient assisted living environments. In addition, the use of depth information does not allow recognizing the identity of people in the scene, thus enabling their detection while preserving their privacy. The proposed DNN has been experimentally evaluated and compared with other state-of-the-art approaches, including both classical and DNN-based solutions, under a wide range of experimental conditions. The achieved results allows concluding that the proposed architecture and the training strategy are effective, and the network generalize to work with scenes different from those used during training. We also demonstrate that our proposal outperforms existing methods and can accurately detect people in scenes with significant occlusions.

Continuous orthogonal moments (COMs) are a type of effective image features widely used in various fields of image processing. However, most of the existing COMs are used for processing flat images and are not suitable for color stereoscopic images. For this reason, this paper first proposes an octonion theory applicable to color stereoscopic images, all color components of color stereoscopic images are coded by using the imaginary part of octonion, and all color components are processed as a whole, and the internal relations among all components are preserved. Then this paper combines the octonion theory with COMs to propose the octonion continuous orthogonal moments (OCOMs). The OCOMs fully reflect and retain the specific correlations between the left- and right-view components of color stereoscopic images, and provide good image description capability. Experimental results show that OCOMs have strong stability and good reconstruction performance when processing color stereoscopic images. Compared with other zero-watermarking methods, the zero-watermarking method embedded by OCOMs has stronger robustness.

This paper presents a general cycle optimization framework with a Kalman filter (KF) module for single-view depth prediction and camera pose estimation. The framework designs a KF module based on measurement noise estimated from networks without supervision to reduce the noise of pose parameters and optimizes the DepthNet architecture to add a new upconvolutional module and a decoder structure to overcome the gradient locality and adjust the mode of multi-task coupling. All modules are integrated to construct a cycle optimization strategy as the core of this paper for overall performance improvement. Experimental results on the KITTI dataset show that the cycle optimization framework greatly improves the performance of the original framework and is better than other improvements on the same original framework; single-view depth prediction and camera pose estimation achieve state-of-the-art performance compared with existing methods under the same or comparable structure.

Fuzzy sets have been widely concerned due it can better deal with more complex and uncertain problems. To expand the application of fuzzy sets, Yager proposed Pythagorean fuzzy sets (PFS) which has been used in many fields. However, in the most cases, the PFS can be generated by expert assessment, which can limit the application of PFS. Hence, how to reasonably generate Pythagorean fuzzy number from known information is a problem worthy of discussion. The paper proposed a method to generate the Pythagorean fuzzy number by considering the negation of probability, namely Pythagorean fuzzy sets based on negation (NPFS). The probability and the probability after negation can be understood as membership and non-membership in the Pythagorean fuzzy number. Besides, NPFS is also set the connection between probability and PFS. There are some numerical examples used to explain the proposed NPFS. In order to explore the effectiveness of NPFS, the paper can not only apply NPFS to artificial decision-making, but also to data-driven applications, which used to handle the multi-criterion decision making (MCDM) and classification problems. In MCDM, the paper applied the NPFS into TOPSIS and used the fault diagnosis to verify its effectiveness by comparing with the results of probability. In the application of classification, the paper proposed the classification method based on NPFS and used real world data to verify the effectiveness of the proposed method.

Text understanding and reasoning are among the core areas of artificial intelligence. Even a total solution to automatic text understanding and reasoning is still beyond the current techniques, thus it is time to build the stepping stones to solve a more straightforward problem set like question-answering (QA) without ambiguous utterances in the contexts or questions. The reported state-of-the-art approaches to this kind of problem are nearly all connectionist models based on neural networks. Significant progress has even been made in this direction. It is still hard for the pure connectionist models to handle logical reasoning. They generally suffer from the longstanding drawback of poor explainability and sensitivity to data noise and distribution. In this paper, we propose a complementary symbolic approach, GMR (Graph Matching based Reasoner) to QA — it automatically generates reasoning rules in the form of graphs from the training set and uses the generated rules to infer answers to the questions in the test set via graph matching. By employing this symbolic approach, 20 tasks in bAbI are solved with an average accuracy of 99.38%, and it outperforms the state-of-the-art for a real-life QA dataset WikiTableQuestions. After analyzing the accuracy of the basic evaluation indicators, we studied the generalization ability of the model in the paper, including the anti-noise ability, the convergence of the model, the stability of the model, the complexity of the algorithm, and the uncertainty of the parameters. Through comprehensive analysis and comparison, our model is stronger than the neural network model regarding anti-noise interference. Compared with the neural network model, our model performs very well in multi-tasking, and the stability of the model is quite high. The diversity of tasks did not reduce the stability of the model. We have conducted a comprehensive analysis and comparison of the parameter uncertainties. Our model can optimally select parameter configurations and will not cause a sharp drop in performance due to the parameter uncertainties. Finally, we describe the complexity of the GMR method and the optimal configuration of its parameters.

Over the last years, researchers have been paying particular attention to scheduling problems integrating production environments and distribution systems to adopt more realistic assumptions. This paper aims to present a new biased random key genetic algorithm with an iterated greedy local search procedure (BRKGA-IG) for open shop scheduling with routing by capacitated vehicles. We propose approximation and exact algorithms to obtain high-quality solutions in acceptable computational times. This paper presents a new integer linear programming model. The proposed integer model has not been addressed in the revised literature. The objective function adopted is makespan minimization, and we use the relative deviation as performance criteria. BRKGA-IG has a new decoding scheme for OSSP-VRP solutions, an intensive exploitation mechanism with an iterated greedy local search procedure, and a restart mechanism to reduce premature population convergence. With these new mechanisms, the extensive computational experience carried out shows that the proposed metaheuristic BRKGA-IG is promising in solving large-sized instances for the new proposed problem, outperforming all other tested methods.

Installing Energy Storage Systems (ESSs) to improve electrical infrastructures of Direct-current (DC) railway systems implies considerable investments that must be assessed carefully. Therefore, it is often necessary to combine detailed railway simulations and decision support mechanisms. Unfortunately, most examples in the literature deal with this topic applying only a single-stage optimization approach: the whole installation is undertaken in a single step, assuming the total budget is available. This paper presents a comprehensive methodology to assess the gradual deployment of the installations when the budget is split into different time periods. This approach is a common situation in real projects and has not been studied yet in the literature. Most often, this type of multi-stage problem is tackled by optimizing each stage independently. On the contrary, this paper proposes to take decisions considering the global impact of each stage optimization, rendering a more efficient solution. This paper proposes a multi-stage formulation of two nature-inspired optimization algorithms (Genetic and Fireworks) to address the installation of ESSs in a realistic railway line. Results demonstrate the excellent behavior of the proposed multi-stage optimization.

Particle filter (PF) is introduced to tackle the limitations of the Kalman filter which adopts Gaussian in the state and noise of the system. PFs have the problem of sample impoverishment and one approach to solve this problem is to optimize the proposal distribution shown by particles. This paper introduces a novel evolutionary PF based on Opposition based High Dimensional optimization Algorithm (OHDA) to reposition the particles of PF in high probable regions for estimation. OHDA will preserve the diversity of particles while emphasizing the more informative ones by information sharing and angular movement operators. Opposite particles are introduced in this paper to speed up the convergence of PF. Virtual forward movement by angular movement of OHDA is employed to better guide the search process. The optimized PF can improve the performance of the estimation algorithms in problems such as localization and SLAM. In robot localization problem, particles show the location of the robot in a known environment. For SLAM (Simultaneous Localization And Mapping), particles contain the location of the robot as well as estimated map of the environment. The application of the resulting evolutionary particle filter is tested in both localization and SLAM. Comparing the results of the proposed evolutionary particle filter with other algorithms confirms the efficiency of applying OHDA to PF in terms of improving estimation accuracy in the well-known Victoria park dataset and some other generated test environments. Comparing optimization algorithms on FASTSLAM and UFASTSLAM are PSO, FA, MVO, and MGWO.

Model predictive control (MPC) is increasingly being considered for control of fast systems and embedded applications. However, the MPC has some significant challenges for such systems. Its high computational complexity results in high power consumption from the control algorithm, which could account for a significant share of the energy resources in battery-powered embedded systems. The MPC parameters must be tuned, which is largely a trial-and-error process that affects the control performance, the robustness and the computational complexity of the controller to a high degree. In this paper, we propose a novel framework in which any parameter of the control algorithm can be jointly tuned using reinforcement learning(RL), with the goal of simultaneously optimizing the control performance and the power usage of the control algorithm. We propose the novel idea of optimizing the meta-parameters of MPCwith RL, i.e. parameters affecting the structure of the MPCproblem as opposed to the solution to a given problem. Our control algorithm is based on an event-triggered MPC where we learn when the MPC should be re-computed, and a dual mode MPC and linear state feedback control law applied in between MPC computations. We formulate a novel mixture-distribution policy and show that with joint optimization we achieve improvements that do not present themselves when optimizing the same parameters in isolation. We demonstrate our framework on the inverted pendulum control task, reducing the total computation time of the control system by 36% while also improving the control performance by 18.4% over the best-performing MPC baseline., This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

In the field of Particle Physics, the behaviors of elementary particles differ among themselves on subtle details that need to be identified to further our understanding of the universe. Machine learning is being increasingly applied in order to solve this task by extracting and extrapolating patterns from detector data. This paper tackles the classification of simulated traces from a liquid argon container into photon- or electron-induced events. Several viable dataset representations are proposed and evaluated on nine supervised learning algorithms to find promising combinations. After that, a hyperparameter optimization step is applied on some of the classifiers to try to maximize their accuracy. Random Forest and XGBoost achieve the best results with roughly 88% test-set accuracy, which shows the potential of machine learning to solve a significant research question in a subfield that is expected to keep growing in the coming years.

With the global outbreak of COVID-19, there is an urgent need to develop an effective and automated detection approach as a faster diagnostic alternative to avoid the spread of COVID-19. Recently, broad learning system (BLS) has been viewed as an alternative method of deep learning which has been applied to many areas. Nevertheless, the sparse autoencoder in classical BLS just considers the representations to reconstruct the input data but ignores the relationship among the extracted features. In this paper, inspired by the effectiveness of the collaborative-competitive representation (CCR) mechanism, a novel collaborative-competitive representation-based autoencoder (CCRAE) is first proposed, and then collaborative-competitive broad learning system (CCBLS) is proposed based on CCRAE to effectively address the issues mentioned above. Moreover, an automated CCBLS-based approach is proposed for COVID-19 detection from radiology images such as CT scans and chest X-ray images. In the proposed approach, a feature extraction module is utilized to extract features from CT scans or chest X-ray images, then we use these features for COVID-19 detection with CCBLS. The experimental results demonstrated that our proposed approach can achieve superior or comparable performance in comparison with ten other state-of-the-art methods.

The control of the pandemic caused by SARS-CoV-2 is a challenge for governments all around the globe. To manage this situation, countries have adopted a bundle of measures, including restrictions to population mobility. As a consequence, drivers face with the problem of obtaining fast routes to reach their destinations. In this context, some recent works combine Intelligent Transportation Systems (ITS) with big data processing technologies taking the traffic information into account. However, there are no proposals able to gather the COVID-19 health information, assist in the decision-making process, and compute fast routes in an all-in-one solution. In this paper, we propose a Pandemic Intelligent Transportation System (PITS) based on Complex Event Processing (CEP), Fuzzy Logic (FL) and Colored Petri Nets (CPN). CEP is used to process the COVID-19 health indicators and FL to provide recommendations about city areas that should not be crossed. CPNs are then used to create map models of health areas with the mobility restriction information and obtain fast routes for drivers to reach their destinations. The application of PITS to Madrid region (Spain) demonstrates that this system provides support for authorities in the decision-making process about mobility restrictions and obtain fast routes for drivers. PITS is a versatile proposal which can easily be adapted to other scenarios in order to tackle different emergency situations.

Loop closure detection helps simultaneous localization and mapping systems reduce map and state uncertainty via recognizing previously visited places along the path of a mobile robot. However, visual loop closure detection is susceptible to scenes with dynamic objects and changes in illumination, background, and weather conditions. This paper introduces PlaceNet, a novel plug-and-play model for visual loop closure detection. PlaceNet is a multi-scale deep autoencoder network augmented with a semantic fusion layer for scene understanding. The main idea of PlaceNet is to learn where not to look in a dynamic scene full of moving objects, i.e., avoid being distracted by dynamic objects to focus on the scene landmarks instead. We train PlaceNet to identify dynamic objects in scenes via learning a grayscale semantic map indicating the position of static and moving objects in the image. PlaceNet generates semantic-aware deep features that are robust to dynamic environments and scale invariant. We evaluated our method on different challenging indoor and outdoor benchmarks. To conclude, PlaceNet demonstrated competitive results compared to the state-of-the-art methods over various datasets used in our experiments.

The importance of considering related stocks data for the prediction of stock price movement has been shown in many studies, however, advanced graphical techniques for modeling, embedding and analyzing the behavior of interrelated stocks have not been widely exploited for the prediction of stocks price movements yet. The main challenges in this domain are to find a way for modeling the existing relations among an arbitrary set of stocks and to exploit such a model for improving the prediction performance for those stocks. The most of existing methods in this domain rely on basic graph-analysis techniques, with limited prediction power, and suffer from a lack of generality and flexibility. In this paper, we introduce a novel framework, called GCNET that models the relations among an arbitrary set of stocks as a graph structure called influence network and uses a set of history-based prediction models to infer plausible initial labels for a subset of the stock nodes in the graph. Finally, GCNET uses the Graph Convolutional Network algorithm to analyze this partially labeled graph and predicts the next price direction of movement for each stock in the graph. GCNET is a general prediction framework that can be applied for the prediction of the price fluctuations of interacting stocks based on their historical data. Our experiments and evaluations on a set of stocks from the NASDAQ index demonstrate that GCNET significantly improves the performance of SOTA in terms of accuracy and MCC measures.

With the wide spread of sensors and smart devices in recent years, the data generation speed of the Internet of Things (IoT) systems has increased dramatically. In IoT systems, massive volumes of data must be processed, transformed, and analyzed on a frequent basis to enable various IoT services and functionalities. Machine Learning (ML) approaches have shown their capacity for IoT data analytics. However, applying ML models to IoT data analytics tasks still faces many difficulties and challenges, specifically, effective model selection, design/tuning, and updating, which have brought massive demand for experienced data scientists. Additionally, the dynamic nature of IoT data may introduce concept drift issues, causing model performance degradation. To reduce human efforts, Automated Machine Learning (AutoML) has become a popular field that aims to automatically select, construct, tune, and update machine learning models to achieve the best performance on specified tasks. In this paper, we conduct a review of existing methods in the model selection, tuning, and updating procedures in the area of AutoML in order to identify and summarize the optimal solutions for every step of applying ML algorithms to IoT data analytics. To justify our findings and help industrial users and researchers better implement AutoML approaches, a case study of applying AutoML to IoT anomaly detection problems is conducted in this work. Lastly, we discuss and classify the challenges and research directions for this domain., Published in Engineering Applications of Artificial Intelligence (Elsevier, IF:7.8); Code/An AutoML tutorial is available at Github link: https://github.com/Western-OC2-Lab/AutoML-Implementation-for-Static-and-Dynamic-Data-Analytics

The successful integration of autonomous robots in real-world environments strongly depends on their ability to reason from context and take socially acceptable actions. Current autonomous navigation systems mainly rely on geometric information and hard-coded rules to induce safe and socially compliant behaviors. Yet, in unstructured urban scenarios these approaches can become costly and suboptimal. In this paper, we introduce a motion planning framework consisting of two components: a data-driven policy that uses visual inputs and human feedback to generate socially compliant driving behaviors (encoded by high-level decision variables), and a local trajectory optimization method that executes these behaviors (ensuring safety). In particular, we employ Interactive Imitation Learning to jointly train the policy with the local planner, a Model Predictive Controller (MPC), which results in safe and human-like driving behaviors. Our approach is validated in realistic simulated urban scenarios. Qualitative results show the similarity of the learned behaviors with human driving. Furthermore, navigation performance is substantially improved in terms of safety, i.e., number of collisions, as compared to prior trajectory optimization frameworks, and in terms of data-efficiency as compared to prior learning-based frameworks, broadening the operational domain of MPC to more realistic autonomous driving scenarios.

The research presented in this paper is aiming to address the safety issue that hearing-impaired people are facing when it comes to identifying a siren sound. For that purpose, a siren identification system, using deep learning, was designed, built, and tested. The system consists of a convolutional neural network that used image recognition techniques to identify the presence of a siren by converting the incoming sound into spectrograms. The problem with the lack of datasets for the training of the network was addressed by generating the appropriate data using a variety of siren sounds mixed with relevant environmental noise. A hardware interface was also developed to communicate the detection of a siren with the user, using visual methods. After training the model, the system was extensively tested using realistic scenarios to assess its performance. For the siren sounds that were used for training, the system achieved an accuracy of 98 per cent. For real-world siren sounds, recorded in the central streets of London, the system achieved an accuracy of 91 per cent. When it comes to the operation of the system in noisy environments, the tests showed that the system can identify the presence of siren when this is at a sound level of up to -6 db below the background noise. These results prove that the proposed system can be used as a base for the design of a siren-identification application for hearing-impaired people.

Physics-Informed Neural Networks (PINNs) have gained much attention in various fields of engineering thanks to their capability of incorporating physical laws into the models. However, the assessment of PINNs in industrial applications involving coupling between mechanical and thermal fields is still an active research topic. In this work, we present an application of PINNs to a non-Newtonian fluid thermo-mechanical problem which is often considered in the rubber calendering process. We demonstrate the effectiveness of PINNs when dealing with inverse and ill-posed problems, which are impractical to be solved by classical numerical discretization methods. We study the impact of the placement of the sensors and the distribution of unsupervised points on the performance of PINNs in a problem of inferring hidden physical fields from some partial data. We also investigate the capability of PINNs to identify unknown physical parameters from the measurements captured by sensors. The effect of noisy measurements is also considered throughout this work. The results of this paper demonstrate that in the problem of identification, PINNs can successfully estimate the unknown parameters using only the measurements on the sensors. In ill-posed problems where boundary conditions are not completely defined, even though the placement of the sensors and the distribution of unsupervised points have a great impact on PINNs performance, we show that the algorithm is able to infer the hidden physics from local measurements., Comment: 22 pages, 46 figures, 6 tables

Visual sensor networks (VSNs) constitute a fundamental class of distributed sensing systems, with unique complexity and appealing performance features, which correspondingly bring in quite active lines of research. An important research direction consists in the identification and estimation of the VSN sensing features: these are practically useful when scaling with the number of cameras or with the observed scene complexity. With this context in mind, this paper introduces for the first time the idea of Stimulation Model (SM), as a mathematical relation between the set of detectable events and the corresponding stimulated cameras observing those events. The formulation of the related SM identification problem is proposed, along with a proper network observations model, and a solution approach based on deep embedded features and soft clustering. In detail: first, the Gaussian Mixture Modeling is employed to provide a suitable description for data distribution, while an autoencoder is used to reduce undesired effects due to the so-called curse of dimensionality emerging in case of large scale networks. Then, it is shown that a SM can be learnt by solving Maximum A-Posteriori estimation on the encoded features belonging to a space with lower dimensionality. Numerical results on synthetic scenarios are reported to validate the devised estimation algorithm., 17 pages, 9 figures, 5 tables, submitted to Engineering Applications of Artificial Intelligence (Special Issue on Intelligent Control and Optimisation organised by IFAC Technical committee: TC3.2 Computational Intelligence and Control)

Evolutionary robotics is concerned with optimizing autonomous robots for one or more specific tasks. Remarkably, the energy needed to operate autonomously is hardly ever considered. This is quite striking because energy consumption is a crucial factor in real-world applications and ignoring this aspect can increase the reality gap. In this paper, we aim to mitigate this problem by extending our robot simulator framework with a model of a battery module and studying its effect on robot evolution. The key idea is to include energy efficiency in the definition of fitness. The robots will need to evolve to achieve high gait speed and low energy consumption. Since our system evolves the robots’ morphologies as well as their controllers, we investigate the effect of the energy extension on the morphologies and on the behavior of the evolved robots. The results show that by including the energy consumption, the evolution is not only able to achieve higher task performance (robot speed), but it reaches good performance faster. Inspecting the evolved robots and their behaviors discloses that these improvements are not only caused by better morphologies, but also by better settings of the robots’ controller parameters.

Image semantic segmentation technology is one of the key technologies for intelligent systems to understand natural scenes. As one of the important research directions in the field of visual intelligence, this technology has broad application scenarios in the fields of mobile robots, drones, smart driving, and smart security. However, in the actual application of mobile robots, problems such as inaccurate segmentation semantic label prediction and loss of edge information of segmented objects and background may occur. This paper proposes an improved structure of a semantic segmentation network based on a deep learning network that combines self-attention neural network and neural network architecture search methods. First, a neural network search method NAS (Neural Architecture Search) is used to find a semantic segmentation network with multiple resolution branches. In the search process, combine the self-attention network structure module to adjust the searched neural network structure, and then combine the semantic segmentation network searched by different branches to form a fast semantic segmentation network structure, and input the picture into the network structure to get the final forecast result. The experimental results on the Cityscapes dataset show that the accuracy of the algorithm is 69.8%, and the segmentation speed is 48/s. It achieves a good balance between real-time and accuracy, can optimize edge segmentation, and has a better performance in complex scenes. Good robustness is suitable for practical application.

In engineering applications, many real-world optimization problems are nonlinear with multiple local optimums. Traditional algorithms that require gradients are not suitable for these problems. Meta-heuristic algorithms are popularly employed to deal with these problems because they can promisingly jump out of local optima and do not need any gradient information. The arithmetic optimization algorithm (AOA), a recently developed meta-heuristic algorithm, uses arithmetic operators (multiplication, division, subtraction, and addition) to solve optimization problems including nonlinear ones. However, the exploration and exploitation of AOA are not effective to handle some complex optimization problems. In this paper, an opposition learning and spiral modelling based AOA, namely OSAOA, is proposed for enhancing the optimization performance. It improves AOA from two perspectives. In the first perspective, the opposition-based learning (OBL) is committed to taking both candidate solutions and their opposite solutions into consideration for improving the global search with a high probability of jumping out of local minima. Then, the spiral modelling is introduced as the second perspective, which is particularly useful in getting the solutions gathering faster and accelerating the convergence speed in the later stage. In addition, OSAOA is compared with other existing advanced meta-heuristic algorithms based on 23 benchmark functions and four engineering problems: the three-bar truss design, the cantilever beam design, the pressure vessel design, and the tubular column design. From our simulations and engineering applications, the proposed OSAOA can provide better optimization results in dealing with these real-world optimization problems.

© 2022 Elsevier LtdLand readjustment and redistribution (LR) is an important approach used to realize development plans by converting rural lands to urban land and also providing urban infrastructure. The LR problem, which is a complex challenging real-world problem, is a discrete optimization problem because its structure is similar to TSP (Traveling Salesman Problem) and scheduling problems which are combinatorial optimization problems. Since classical mathematical methods are insufficient for solving NP (Nondeterministic Polynomial) optimization problems due to time limitations, meta-heuristic optimization algorithms are commonly utilized for solving these kinds of problems. In this paper, meta-heuristic algorithms including genetic, particle swarm, differential evolution, artificial bee, and tree seed algorithms are utilized for solving LR problems. The stated meta-heuristic algorithms are used by applying spatial-based crossover and mutation operators depending upon the LR problem on each algorithm. Moreover, a synthetic dataset is used to ensure that the quality of the solution obtained is acceptable to everyone, to prove an optimal solution easily. By utilizing the suggested spatial-based crossover and mutation operators, finding the ideal solution is aimed using the synthetic dataset. In addition, five different modifications on TSA (Tree-Seed Algorithm) are performed and used to solve LR problems. All the modified versions of TSA are carried out only by changing the mechanism of seed reproduction. The novel TSA approaches are respectively named as tcTSA (tournament current), tbTSA (tournament best), pbTSA (personal-best based), t2TSA (double tournament), and elTSA (elitism based). In the experimental studies, the hybrid approach, which includes the crossover and mutation operators, is successfully applied in all of the algorithms under equal conditions for a fair comparison. According to experimental results performed using the dataset, it can be clearly stated that especially t2TSA outperforms all the algorithms in terms of performance and time.

This paper presents a coupled, neural network-aided longitudinal cruise and lateral path-tracking controller for an autonomous vehicle with model uncertainties and experiencing unknown external disturbances. Using a feedback error learning mechanism, an inverse vehicle dynamics learning scheme utilizing an adaptive Radial Basis Function (RBF) neural network, referred to as the Extended Minimal Resource Allocating Network (EMRAN) is employed. EMRAN uses an extended Kalman filter for online learning and weight updates, and also incorporates a growing/pruning strategy for maintaining a compact network for easier real-time implementation. The online learning algorithm handles the parametric uncertainties and eliminates the effect of unknown disturbances on the road. Combined with a self-regulating learning scheme for improving generalization performance, the proposed EMRAN-aided control architecture aids a basic PID cruise and Stanley path-tracking controllers in a coupled form. Its performance and robustness to various disturbances and uncertainties are compared with the conventional PID and Stanley controllers, along with a comparison with a fuzzy-based PID controller and an active disturbance rejection control (ADRC) scheme. Simulation results are presented for both slow and high speed scenarios. The root mean square (RMS) and maximum tracking errors clearly indicate the effectiveness of the proposed control scheme in achieving better tracking performance in autonomous vehicles under unknown environments.

Several learning-based control with computational intelligence strategies handle challenges related to the difficulty of modeling complex systems or the need for control strategies with provably safe. In recent years, learning-based control using machine learning has been successfully demonstrated in robotics applications and applied to deal with nonlinearities. These control methods may lead to better solutions to nonlinear problems, such as the safety-critical industry, which requires strong guarantees about the controller behavior. Learning-based neural network control can comprehend and learn about plants, disturbances, the environment, and operating conditions. In this paper, we presented a Lyapunov-based nonlinear control determined from a deep neural network, which uses the Lyapunov theory to compute a control law for a nonlinear system. For advance stability analysis, an estimation of the region of attraction is presented. A numerical example and experimental simulations using the rotational inverted pendulum system are performed and compared with a conventional control technique. The proposed method calculated a control law that provided the stabilizability of the system and produced better solutions considering different tracking and process disturbance.

Generalized evidence theory (GET) is a generalization of Dempster–Shafer evidence theory. It copes with information in an open world, which makes up for the shortcoming that Dempster–Shafer evidence theory cannot handle information conflict effectively. However, GET also faces an unavoidable problem: how to determine the number of unknown targets in the incomplete frame of discernment (FOD). Fuzzy C-means (FCM) is a clustering algorithm that divides the original data set into different clusters and summarizes similar data into the same cluster. Therefore, determining the number of unknown targets in the open world can be transformed into finding the number of clusters. However, FCM has the disadvantage of subjectively controlling the number of clusters. In order to overcome this shortcoming, we use fuzzy gap statistic algorithm (FGS) to optimize it. FGS can effectively determine the optimal number of clusters in FCM. Therefore, this paper proposes a new method based on FGS to determine the number of unknown targets in the open world. In addition, to verify the method’s accuracy, we conducted seven experiments based on the University of California Irvine ( UCI ) data sets, including Iris, glass, Haberman, Knowledge, Robot, seeds, and WDBC. Finally, the experimental results illustrate that the proposed method to determine the number of unknown targets in the incomplete FOD has high effectiveness.

Multi-Agent systems technology can offer valuable tools to develop applications in domains involving transactions of assets. However, they usually do not have a proper support for reliable and decentralized recording of the transactions that are common in this kind of system. This support can be provided by the Blockchain technology. Furthermore, it is important to have means to represent in the system concepts that are intangible, such as asset and ownership. This paper presents a model of integration between Multi-Agent Systems and Blockchain where an artificial institution connects the intangible concepts related to transactions of assets to the concrete elements composing the system. An application example illustrates an implementation following the proposed integration model, showing its advantages and limitations. In the example, agents contract each other to provide services upon the payment of a dealt value through a system based on cryptocurrencies and blockchain. It highlights the essential contributions of the proposed approach to systems where agents transact assets: regulation of the system, representation of the notion of asset that does not depend on agents, and reliable recording of transactions based on Blockchain.

The scientific and reasonable prediction of wastewater discharge is of great significance for regional water environment management and water resources protection. To this end, a new high-performance grey prediction model suitable for wastewater discharge prediction named FDGM(1,1, k , r) is proposed based on the three-parameter discrete grey prediction model (TDGM(1,1) for short) in this paper. Firstly, the mechanism and structure defects of TDGM(1,1) are systematically analysed and made up in FDGM(1,1, k , r) by adding a nonlinear correction term and new grey generation operator with real number field ( r ∈ R ). Then, the new information priority principle (Metabolic Thought) is introduced into the new model according to the dynamic nature of wastewater discharge prediction. Thirdly, the empirical results of wastewater discharge in Chongqing show that the mean relative percentage error of new model is only 0.216%, which is superior to other mainstream grey forecasting models of wastewater. Lastly, the new model is used to forecast the wastewater discharge in Chongqing China, and the prediction results show that the wastewater discharge in Chongqing will be as high as about 3.1 billion tones in Year 2025, and the government should formulate timely countermeasures to deal with the rapidly increasing wastewater discharge in the future.

With the boom of big data, the Internet contains more and more personal behavior information, but it is difficult to extract effectively. A model involving multivariate processing capability must be constructed to deal with these time series with complex characteristics. In this paper, a novel hybrid model embedding Baidu Search Index is therefore proposed to implement multi-step ahead subway passenger flow forecasting. Firstly, we collect data from informative Baidu Search Index, reduce dimensionality, and screen out the powerful predictors by statistical analysis. Secondly, we extract matching common modes at similar time scales between the subway passenger flow and screened Baidu Search Index via multivariate mode decomposition being optimized by multi-objective algorithm. Furthermore, to eliminate pseudo statistical causality, we select the optimal combination of modal components between subway passenger flow and its corresponding Baidu Search Index at each time scale by an innovative multi–modal analysis strategy. Thirdly, we reconstruct the forecasting values of each selected optimal combination as the final results. The empirical results of Beijing, Shanghai and Guangzhou show that the proposed model can significantly outperform six benchmark models in both the level and directional accuracy. So introducing Baidu Search Index creates a sound opportunity to enhance the subway passenger flow forecasting ability.

In this paper, we present a constrained Dirichlet process mixture model with labels as side information. Specifically, the labeled information is incorporated through the use of a product partition prior to give clusters of instances with similar labels a higher prior preference. The proposed formulation is further extended to handle multiple side information. The empirical results on several benchmark datasets show that our method can consistently improve its clustering performance as more labeled data become available. Even in the presence of noisy labels, the proposed method rarely performs worse than its unsupervised counterpart. The effectiveness of the proposed method is also demonstrated through an application of magnetic resonance imaging for identifying major brain tissues.

Manual analysis of electrocardiogram (ECG) signals is a laborious and prone-to-error task, even for a specialist with many hours of experience. For this reason, research on automatic ECG diagnosis is widespread in the literature and continues to grow each year. The present paper describes a novel and fully functional expert system for automatic diagnosis of 13 different diseases using standard 12-lead ECGs. This system makes three significant contributions to the state of the art: (a) the large number of different diseases diagnosed; (b) the use of 5 leads for a more precise identification and measurement of the ECG waves; and (c) a novel noise indicator that measures the quality of the acquired ECG signal. The kernel of the system consists of a set of rules that replicate a specialist’s diagnostic process but with the speed of an automatic system. The rules use a set of parameters generated after a noise-filtering process of the ECG signal and subsequent identification of its different waves (P, QRS complex, T, and Delta). The design of the rules was carried out with the collaboration of a specialist with more than 20 years of experience in ECG diagnosis and using a database of 284,000 ECGs as support. The system was validated by the specialist, obtaining a reliability of 80.8%. Given the complexity of the problem and the number of diagnoses covered, the results are considered satisfactory and make the system a useful support tool for diagnosis.

Sine Cosine Algorithm (SCA) that is one of the population-based metaheuristic optimization algorithms basically consists of the updating mechanism based on sine and cosine functions. In this algorithm, a few random and adaptive variables are also utilized for more effective motions of the candidate solutions. SCA has some drawbacks like other some metaheuristic algorithms. SCA tends to be stuck into the local regions in the search space and this affects negatively on the computational effort required to find the best solution point in the search space. This paper presents four different improved versions of SCA. The proposed improvements on original SCA are the innovations on the updating mechanism of SCA. To evaluate the performances of Improved Sine Cosine Algorithms (ImpSCAs), well-known numerical optimization problems including CEC 2014 test suite are used. Firstly, different analyses of the proposed ImpSCAs are dealt with such as the convergence analysis, search history analysis, trajectory analysis, average distance analysis, and computational complexity analysis. Secondly, the proposed four versions of ImpSCAs are compared with the original SCA for CEC 2014 benchmark problems with dimension sizes of 10D, 30D and 50D. Finally, original SCA and ImpSCAs are adapted to select optimal feature combination and they are tested for 10 feature selection datasets taken from the UCI machine learning repository. The benchmark results show that the performances of the ImpSCA 1, ImpSCA 2, and ImpSCA 4 are better than that of the original SCA. From the feature selection results, it is observed that three versions of ImpSCAs (except ImpSCA 3) outperform the original SCA in 80% of the datasets. Source codes of ImpSCAs are publicly available at https://github.com/uguryuzgec/ImpSCAs .

As an important technique for dealing with transaction database in the field of data mining, utility-driven mining can be used to discover useful patterns (i.e., itemsets, sequences) which have a high utility. However, it has a bias towards the item/object combinations which have more exhibition period since they have more opportunity to generate a high utility. To address this, the on-shelf time period of items need to be considered, thus on-shelf utility mining (OSUM) can be applied in the application which is more closer to the actual situation. Currently several models have been proposed to deal with the OSUM problem, but they still suffer from the requirement that it needs to maintain a massive candidates in memory and to scan database many times. In this paper, we propose two effective one-phase algorithms named OSUMI (On-Shelf Utility Mining from transactIon database) and OSUMI + (the improve version of OSUMI). Both OSUMI and OSUMI + search all itemsets as a set-enumeration tree and discover the on-shelf itemsets with high utility in a more practical way. More precisely, in order to avoid the problems of high memory consumption, two algorithms apply some properties of the concept of on-shelf utility. Besides, two upper-bounds named subtree utility and local utility are applied to early filter out unpromising patterns and then prune the search space. Finally, an extensive experimental study on several real on-shelf datasets shows that our proposed algorithms can be significantly faster than the state-of-the-art algorithm.

In real-world applications, there are many multimodal multi-objective optimization problems, which have multiple equivalent global Pareto-optimal solutions or with at least one local Pareto-optimal solution in the decision space. While some evolutionary algorithms have been proposed to find the global solutions recently, they are difficult to handle multimodal multi-objective optimization problems with local solutions. Meanwhile, there have been few studies on searching for local Pareto solutions. However, local solutions are additional alternatives for the decision makers if global solutions are impracticable. This paper proposes a particle swarm optimizer based on reference point, termed RPPSO, which combines a reference point mechanism and a local solution preserving technique. The reference point strategy is utilized to establish multiple evenly distributed neighborhoods and guide particles to evolve independently in their respective neighborhoods, so as to detect more Pareto solutions in the decision space. The local solution preserving technique is employed to estimate the dominant radius of each front, and with this radius to classify the individuals as either non-local or local solutions with the aim of retaining the local solutions. In addition, a set of benchmark test functions with local Pareto solutions are designed. The proposed algorithm is comprehensively evaluated on forty-four benchmark functions and is compared with fourteen state-of-the-art algorithms. The experimental results show that the proposed RPPSO achieves competitive performance than its competitors in terms of the reciprocal of Pareto sets proximity (rPSP). The RPPSO is also applied to solve on one real-world problem (i.e., map-based problem) to further verify the effectiveness and efficiency.

Intuitionistic fuzzy set (IFS) is a classical branch of fuzzy set, which has advantage to deal with uncertain problems. In IFS, similarity measure is an important fundamental concept, it is used to measure consistency between different intuitionistic fuzzy sets (IFSs) and becomes a key parameter in fuzzy decision system. However, the previous methods of similarity measure do not take enough account the effect of hesitancy degree on membership degree and non-membership degree, so that produce counterintuitive results when measuring similarity. Hence, in this paper, a new similarity measure of IFS is presented. The effect of hesitancy degree on similarity measure is fully considered in the proposed method and some properties also haven been discussed to prove the reasonable of proposed method. Meanwhile, some numerical examples are analyzed to illustrate characteristics of proposed similarity measure in detail. Further, the experiments of target classification and clustering problem demonstrate effectiveness and superiority of proposed similarity measure in the environment of expert assessments and data set.

Truck-drone hybrid delivery is a hybrid one combining the advantages including large capacity of truck and high travel speed of drone together. Vehicle routing problem with drones (VRP-D) is a common one in the above delivery system. In this paper, VRP-D is addressed and a new dynamical artificial bee colony (DABC) is employed to minimize the overall operational cost. Two bee swarms are produced and an effective evaluation process is used to determine employed bee swarm and onlooker bee swarm dynamically. Variable neighborhood descent is constructed by using 15 neighborhood structures and adopted in employed bee phase and onlooker bee phase in different ways. A number of experiments are conducted on 112 instances and the computational results reveal that DABC provides new best solutions for 37 instances and has promising advantages on VRP-D.

As an excellent discriminant classifier based on generating prior knowledge, the minimax probability machine (MPM) has been widely used and deeply researched in many fields. The core idea of minimax probability machine is to directly estimate probability accuracy bound by minimizing the maximum probability of misclassification. However, minimax probability machine does not include a regularization term for the construction of the separating hyperplane, and it needs to solve a large-scale second-order cone programming problem in the solution process, which greatly limits it development and application. In this paper, to improve the performance of minimax probability machine, we propose a novel binary classification method called regularized twin minimax probability machine classification (TMPMC). The TMPMC constructs two non-parallel hyperplanes for final classification by solving two smaller second-order cone programming problems to improve the performance of the MPM. For each hyperplane, our method is theoretically well grounded on the idea of minimizing the worst case (maximum) probability of misclassification of a class of samples while the distance to the other class is as large as possible. Our approach was first derived as linear methods, and subsequently extended as kernel-based strategies for nonlinear classification. Additionally, we extend TMPMC to the regression problem and propose a new regularized twin minimax probability machine regression (TMPMR). Experimental results on several datasets show that our methods are competitive in terms of generalization performance compared to other algorithms.

The distillation unit (DU) is an essential product separation unit in refineries. The process operation of DU is directly related to the quality and yield of the final petroleum products. The DU studied in this work is deeply troubled by the varying feedstock properties, which aggravates the difficulty of process operation. To determine the proper operation variables, a knowledge-based operation optimization (KOO) strategy of a DU is proposed in this paper. The KOO strategy is composed of a supervision module and an optimization module. First, the operating conditions are divided into four types based on the feedstock properties. In supervision module, an improved bar-shaped convolutional neural network supervision model (IBS-CNN-based SM) is developed to monitor the operating conditions. The model output which represents the current operating condition information is transmitted to the lower optimization module. In optimization module, the fuzzy-logic-based optimization strategy is designed to adjust two temperature variables — the top temperature of the distillation column (TTDC) and the outlet temperature of the re-boiling furnace (OTRF) to ensure the product quality requirements. Industrial experiments have illustrated the KOO strategy could adapt to the varying feedstock properties. During the experiment, the proposed KOO strategy improved the product qualification rate from 86.67% to 93.34% and saved the consumption of gas and cooling water to a certain extent.

The digital revolution of noncontact physiological signal monitoring in clinical and home health care is underway, and deep learning techniques are incredibly popular. Camera-based physiological signal monitoring for adults has made considerable progress in recent years. However, most of existing methods and datasets are developed for adult subjects, and until now, there has been no neonatal public database that is collected for developing deep learning method. Thus, in this paper, we introduce a large-scale newborn baby database, named NBHR (newborn baby heart rate estimation database), to fill the abovementioned knowledge gap. A total of 9.6 h of clinical videos (1130 videos totaling 921 GB) and reference vital signs are recorded from 257 infants at 0–6 days old. The facial videos and corresponding synchronized physiological signals, including photoplethysmograph information, heart rate, and oxygen saturation level, are recorded in our database. This large-scale database could be used to develop deep learning methods to estimate heart rate or oxygen saturation levels. Furthermore, a multitask deep learning method, called NBHRnet, is proposed to estimate heart rate based on the NBHR database, and the model is succinct that it can be deployed on a computer without GPUs. The experimental results indicate that NBHRnet yields competitive performance in predicting infant heart rate, with a mean absolute error of 3.97 bpm and a mean absolute percentage error of 3.28%; additionally, it can estimate heart rate almost instantaneously (2 s/60 frames). Our datasets are freely publicly available by request.

Automated test case generation for path coverage (ATCG-PC), as an important task in software testing, aims to achieve the highest path coverage of a tested program by using as little computational overhead as possible. In ATCG-PC, “similar paths are usually executed by similar test cases” is a problem-specific knowledge which was touched by a handful of researchers but still underutilized. Inspired by the problem-specific knowledge, this paper designs a local search strategy by improving a scatter search strategy, and then proposes a grey prediction evolution algorithm with the improved scatter search strategy for ATCG-PC. Here, the improved scatter search strategy could obtain two feasible test cases by exploiting a dimension of a test case covering a certain path. The proposed algorithm is constructed by importing the improved scatter search strategy to the end of the reproduction operation of the grey prediction evolution algorithm holding strong exploration ability. Grey prediction evolution algorithm is first applied to solve ATCG-PC. The performance of the proposed algorithm is evaluated on six fog computing benchmark programs and six natural language processing benchmark programs. The experimental results demonstrate that the proposed algorithm can achieve the highest path coverage with the fewer test cases and running time than some state-of-the-art algorithms.

The three-parameter (3-p) Weibull distribution is an extremely important distribution to characterise the statistical behaviour of a large number of real world phenomenons. It is also useful as a failure model in analysing the reliability of different types of mechanical and electrical components/systems. Successful applications of the distribution rely on an accurate estimation of its three parameters because it directly affects the reliability and lifetime analysis. Due to the intricate system of nonlinear equations and the complexity of the likelihood function, derivative-based optimisation methods may fail to converge. Thus, an efficient and effective method for estimating the parameters of the model is important from the practical viewpoint. In this paper, an optimisation scheme based on an acceptance–rejection (AR) mechanism coupled with an elegant nested sampling (NS) technique is proposed to tackle this problem. The idea is to gradually approach the region of optimal solutions through an efficient sampling technique and a reweighting scheme. The AR-NS algorithm allows a good exploration of the parameter space and converges towards higher likelihood regions by decreasing progressively a pre-specified tolerance threshold. The proposed approach gives the entire distributions of the optimal estimates rather than a single point estimates. To demonstrate the practicality and the efficiency of the proposed approach, numerous numerical examples using simulated data and real-world engineering cases will be given. The obtained results show that the AR-NS algorithm is a suitable method for estimating the parameters of lifetime distributions using different distances.

This paper presents a new method for initializing weights in a Feedforward Neural Network (FNN) with a single hidden layer combined with a constructive approach to define the number of hidden units associated with the best classification performance. The strategy consists of defining an initial number of hidden units according to the classification problem, the linearization of the whole network around an equilibrium point and the determination of the initial weights and bias through the maximum approximation of the linearized model to the Optimal Linear Classifier (OLC) whose solution can be obtained analytically. The constructive algorithm comprises a gradual increase in the number of hidden units in such a way that at each training only the weights and bias associated with the new hidden units are initiated randomly while the weights and bias obtained from previous training are used as initial guesses. Additionally, the constructive algorithm seeks to ensure that the loss function of the trained networks decreases with the successive additions of hidden units. The proposed approach (Weight Initialization based on the Linearization of the Whole Neural Network combined with a new Constructive Algorithm, WILWNN-CA) is applied to synthetic and real datasets widely used as benchmark for multi-class classification problems. The comparison with conventional random weight initialization and other approaches involving different network topologies (and initialization strategies) shows that the proposed method is efficient and capable of providing success rates (correct classification rates) higher or similar to those achieved with existing methods.