Search

Risks are uncertain events that can affect criteria such as project cost, quality, and completion time and ultimately lead to project failure. That is why the project risk management process, which is one of the most fundamental parts of project management, must be done properly. In this paper, a new multi-objective fuzzy model is proposed to respond appropriately to the primary and secondary project risks. The contributions, such as risk interdependency, considering the project with multi-mode activities, resource considerations, as well as attention to interval-valued fuzzy uncertainties are regarded simultaneously for the first time. The purpose of the proposed multi-objective mathematical model is to minimize cost, quality reduction, and project completion time. Then, a new two-stage uncertain solution approach is introduced in this paper. In this solution approach, using an equivalent method in the first step and then using an extended multi-choice goal programming method in the second step, for both lower and upper limits of the membership functions for interval-valued fuzzy numbers, the computational processes are performed. To show the application and effectiveness of the proposed model, a case study adapted from the literature is given. Finally, sensitivity analysis is conducted to analyze the behavior of the developed mixed-integer programming model. The analysis of obtained results indicates that the utilization of the proposed model leads to better and more appropriate solutions. • Presenting a new multi-objective programming model for project risk responses. • Proposing multi-mode activity and resource considerations in project risk modeling. • Introducing a new interval-valued fuzzy solution approach to handle uncertainty. • Focusing on primary and secondary risks along with project risk interdependency. • Implementing the proposed model in a case study for electricity industry projects. [ABSTRACT FROM AUTHOR]

The strategic management of reverse logistics (RL) is essential for enterprises to enhance their operational efficiency, customer satisfaction, and sustainability performance in today's competitive marketplace. Many manufacturing firms have to cooperate with professional RL providers to overcome resource constraints and technological limitations, ultimately driving business success. Thus, it is critical for every enterprise to select the most suitable third-party RL provider (3PRLP). This paper provides a novel group decision making (GDM) method with incomplete triangular fuzzy multiplicative preference relations (TFMPRs) to cope with the selection of the most optimal 3PRLP. Firstly, a definition of acceptable incomplete TFMPRs is given. Then, the sufficient and necessary condition of an acceptable incomplete TFMPR is proposed. By analyzing the properties of consistent TFMPRs, a graph-based algorithm is designed to estimate the unknown elements in incomplete TFMPRs. Based on the proposed acceptable consistency definition of TFMPRs, an optimization model is set up to improve the consistency degree of inconsistent TFMPRs. The optimal normalized triangular fuzzy multiplicative weight vector (Tri-MWV) is obtained by computing two analytic expressions and solving a linear programming model. To measure the closeness degree of two TFMPRs, the concept of logarithmic correlation coefficient (LCC) between two TFMPRs is proposed. Combining the incomplete preference information in incomplete TFMPRs with the LCC s of any two TFMPRs, an algorithm of computing experts' weights is displayed. Subsequently, a novel method of GDM with incomplete TFMPRs is presented. Lastly, a practical example of evaluating 3PRLPs is conducted to illustrate the effectiveness of the proposed GDM method with incomplete TFMPRs. • A novel group decision making with incomplete triangular fuzzy multiplicative preference relations for the selection of an optimal third-party reverse logistics provider. • Give a concept of acceptable incomplete triangular fuzzy multiplicative preference relations. • Design a graph-based algorithm to estimate the unknown elements in TFMPRs. • Obtain the optimal normalized triangular fuzzy multiplicative weight vector. • A method of determining expert's weight vector is put forward. [ABSTRACT FROM AUTHOR]

Synchronous motors play a significant role in a wide range of industrial applications. Their reliable operation is paramount. Any faults in synchronous motors can lead to costly downtime, decreased productivity, and potential safety hazards. By accurately diagnosing and classifying faults, we can proactively address issues before they escalate, ensuring the smooth operation of synchronous motors and minimizing the risk of equipment failure. The accurate diagnosis and fault detection in synchronous motors pose a significant challenge in their operation and maintenance. In the existing models, the feature data at various depths are not thoroughly extracted to maximize their feature extraction capability. Additionally, they employ a single support vector machine to make the final decision on the output. The single support vector machine may not consistently produce more accurate outcomes. Therefore, this paper proposes a novel fault diagnosis model based on a deep residual neural network and multiple support vector machines to diagnose mechanical and electrical faults of synchronous motors. The proposed model improves upon existing fault diagnosis models in two key aspects. Firstly, by employing a deep neural network, the model is able to effectively process and extract fault features from the motor fault dataset, capturing more nuanced information that may be missed by existing models. Secondly, the use of multiple support vector machines enhances the decision-making capability of the model, allowing for more accurate fault diagnosis. By combining these two aspects, the proposed model achieves superior diagnostic performance compared to single support vector machine-based models. Our proposed model has been rigorously evaluated using mechanical and electrical fault datasets, and the results of experimental tests clearly demonstrate its superior diagnostic performance when compared to existing fault diagnosis models. The synergy of deep neural networks and multiple support vector machines not only improves fault detection accuracy but also enhances the robustness and generalizability of the model, making it a valuable tool for real-world industrial applications. • A novel model is proposed for synchronous motor fault diagnostics. • A deep residual neural network effectively extracts fault features. • Multiple support vector machines enhance the detection capability. • Proposed model is compared with existing models. [ABSTRACT FROM AUTHOR]

Human-Inspired Control (HIC) holds promise for endowing machines with human-like cognition, decision-making, and adaptability. In this study, we employ a fusion of cognitive modeling, machine learning, and control theory as the foundational architecture and empirically validate its suitability for robot control within the realm of HIC. Fuzzy logic stands out as a viable approach for HIC control, wherein control rules can be devised drawing from human intuitive inspiration. Specifically, this study explores similarity inference control systems in robotics, with the objective of enhancing line-following control as an alternative to fuzzy systems. The experimental optimization results provide insights into the advantages and limitations of the similarity inference control system. Despite achieving performance comparable to that of traditional two-stage fuzzy control systems, careful consideration of noise sensitivity is paramount. While the similarity inference approach streamlines implementation and obviates the necessity for expert-designed fuzzy rules, its susceptibility to noise can compromise performance, particularly in noisy environments. These considerations are pivotal for the development of control systems aimed at mitigating noise sensitivity, enhancing task-specific performance, and ensuring the adaptability and robustness of line-following robots. To tackle this challenge, we both discuss and experimentally evaluate potential solutions and their applicability in this paper. • The study focuses on integrating cognitive modeling, machine learning, and control theory in HIC, exploring methods like fuzzy logic. • It delves into robotics' similarity inference control systems, optimizing line-following, tackling noise susceptibility, and suggesting enhancements via sensor data and machine learning. [ABSTRACT FROM AUTHOR]

Topological data analysis (TDA) is a powerful mathematical framework that extracts valuable insights about the shape and structure of complex datasets by identifying and analyzing underlying topological features, including connected components, holes, voids, and higher-dimensional counterparts. TDA achieves this by constructing a simplicial complex from the data, comprising simple shapes like points, lines, triangles, and higher-dimensional counterparts. The Vietoris–Rips complex (VRC), a widely used method, constructs simplicial complexes by measuring the distances between data points in a metric space. While the link between Intuitionistic Fuzzy Sets (IFSs) and TDA is clear through intuitionistic fuzzy distance measures, the literature has not explored the application of TDA for extracting topological features from uncertain and vague datasets using IFSs. In this paper, we propose a novel approach that leverages Intuitionistic Fuzzy Distance Measures (IFDMs) to generate the Intuitionistic Fuzzy Vietoris–Rips Complex (IFVRC) of IFSs. Specifically, we investigate the persistence of invariant relations of topological features between IFVRCs constructed using different distances, including Atanassov's Euclidean distance (l 2 I F S), Boran and Akay's distance (D), and Liu's distance measure (d L), over artificially generated IFSs. Furthermore, we demonstrate that the generation of IFVRC can be viewed as a Natural Hierarchical Clustering Algorithm (NHCA) for IFSs. Finally, we apply the proposed IFVRC as a Natural Hierarchical Clustering Algorithm to cluster intuitionistic fuzzy datasets related to Car and Building materials. Moreover, we incorporate a relatively large dataset, the GTZAN dataset from the Kaggle datasets repository, to classify music genres based on their topological information. • Topological analysis of intuitionistic fuzzy distance measures (IFDMs) have been conducted. • Utilizing IFDMs, Intuition Fuzzy Vietoris Rips Complex (IFVRC) generated. • It is shown how to extract topological features from IFVRC. • The persistence relations of topological features between IFVRCs have been investigated. • Finally, IFVRC is treated as a natural hierarchical clustering algorithm. [ABSTRACT FROM AUTHOR]

As an extension of partition, fuzzy β -covering can provide a more realistic and accurate description of incomplete information. In this paper, we mainly integrate the idea of fuzzy β -covering with dual hesitant fuzzy (DHF) information and construct some novel three-way decision (3WD) models with DHF covering-based rough set. Firstly, we propose the notions of DHF β -covering, DHF β -minimal description and DHF β -maximal description, and then construct three types of DHF neighborhood operators. Meanwhile, we introduce DHF conditional probability by using DHF neighborhood operator. Secondly, in terms of DHF conditional probability, DHF covering-based probabilistic rough set, DHF covering-based decision-theoretic rough set and their 3WD models are established. In light of DHF neighborhood operator, we propose two pairs of optimistic and pessimistic DHF decision evaluation functions and build two 3WD models through them. Lastly, a numerical example is employed to elaborate the application of the above models, which is effective and credible to medicine screening for Alzheimer's disease. • Proposed neighborhood operators in dual hesitant fuzzy environment. • Three-way decision based on dual hesitant fuzzy probabilistic rough sets. • Three-way decision based on dual hesitant fuzzy decision-theoretic rough sets. • Three-way decision based on dual hesitant fuzzy decision evaluation functions. • Experiment on medication screening of Alzheimer's disease are given. [ABSTRACT FROM AUTHOR]

The paper presents the application of neural networks evolving under the Hill Climb Assembler Encoding (HCAE) algorithm to control follower autonomous underwater vehicles that are members of a swarm consisting of one leader vehicle and a group of followers. The leader is responsible for global navigation and guiding the swarm, while low-cost followers unable to cover longer distances on their own follow the leader. To locate the leader, the followers only use information about the distance to it. Directional information is unavailable to the followers. Moreover, information about the distance is transmitted to followers with a frequency depending on the number of followers: the more followers, the lower the frequency. In addition to tracking the leader, the followers must also be able to avoid collisions with other followers and the leader. To this end, they are equipped with short-range sensors looking around each follower. The simulations presented in the paper were carried out for different swarm sizes, different sensor ranges, different collision distances, and variable leader speed, and showed high effectiveness of the proposed neural solution. The most common strategy for the followers when following a leader was to circle it at a certain safe distance depending on the sensor range. • Neuro-evolutionary controller for a swarm of autonomous underwater vehicles is proposed. • The task of the controller is to keep the swarm in one compact group while moving along a predefined trajectory. • The only information provided to the controller is the distance to the swarm leader, directional information is unavailable. [ABSTRACT FROM AUTHOR]

As social media platforms grow rapidly, multi-modal data is becoming more and more prevalent. A user can better understand events by analyzing multimodal data for topics. Automatic topic detection from multimodal data can potentially have tremendous value for advertising and government agencies for whom public opinion matters for strategic decisions and policy making. However , multimodal topic detection is complicated for two reasons: (1) The nature of the multimodal data varies from one medium to another, and (2) The noisy nature of webdata. Conventional topic models are ineffective in dealing with these two problems. This paper proposes, a framework for multimodal topic modeling for social media data that uses topics extracted using Latent Dirichlet Allocation (LDA) and patterns found from images using transfer learning. The proposed framework makes use of textual as well as visual data for topic detection. The experiments are conducted on the benchmark datasets: Flickr8k, Flickr30k, and MCG WEBV. The proposed work outperformed other techniques in terms of accuracy (0.63), precision (0.75), recall (0.97), F-Measure (0.85), Bleu-1(0.68),METEOR (0.17) , ROUGE-L (0.49), and CIDEr (0.573). The proposed work is compared to state-of-the-art methods to demonstrate its accuracy. • A multimodal topic modeling framework is proposed using deep transfer learning. • Various deep CNNs architectures are explored & deployed for multimodal topic modeling • Proposed method has been compared with state-of-the-art using three datasets. [ABSTRACT FROM AUTHOR]

For warehouse operations, efficiently scheduling the available resources is crucial to improve the productivity and customer satisfaction. This paper proposes a simulation-based evolutionary algorithm for order scheduling and multi-robot task assignment in a robotic mobile fulfillment system. The algorithm proactively deals with the effects of the processing time variability by evaluating schedules based on both its system performance as well as its robustness under uncertain conditions. The algorithm implements an efficient resource allocation method and a variance reduction technique to reduce the overall computational burden. The experimental results show that the techniques to reduce the computational time are effective and can significantly reduce the amount of simulations required for the fitness evaluation. If a candidate schedule is allocated insufficient simulation replications it can lead to an inaccurate estimate of its long-term average performance. This could lead to an average performance loss of 7.3 %. Furthermore, the proactive scheduler is able to generate schedules that are more robust compared to deterministically generated. A reduction in the average operational cost of about 5 % can be reached, compared to a deterministically generated schedule. The paper reveals the relevance of identifying and modeling uncertainty when designing schedules in an operational system, rather than looking for optimal schedules for ideal scenarios. • We present a proactive evolutionary algorithm for the stochastic RMFS scheduling problem. • We consider the integrated scheduling problem with processing variability. • We implement efficient evaluation, resource allocation and variance reduction procedures. • We evaluate the performance of the algorithm on instances with increasing due date tightness. • The proactive algorithm can generate robust and good quality schedules under uncertainty. [ABSTRACT FROM AUTHOR]

Reinforcement Learning (RL) has gained significant popularity in recent years for its ability to solve complex control problems. However, most existing RL algorithms are designed to train policies for each environment in isolation, limiting their applicability to real-world scenarios with many related environments. Recently, many multitask optimization algorithms have been proposed and successfully applied to a wide range of optimization problems. However, existing studies mainly focus on solving multiple continuous function benchmarks and overlook the potential of tailoring towards RL. In this paper, we propose a simple multitask optimization algorithm called Multitask Augmented Random Search (MARS) that trains multiple RL agents together and exploits the performance surplus from highly correlated tasks. MARS is a modification of the simple random search Augmented Random Search (ARS) algorithm, which has been shown to outperform complicated methods in solving continuous control MuJoCo environments such as Soft Actor-Critic (SAC), Proximal Policy Optimization (PPO), and Trusted Region Policy Optimization (TRPO). The experimental results also demonstrate that our proposed algorithm is more consistent in solving different instances of MuJoCo benchmark than ARS, Multifactorial Evolution Algorithm (MFEA), and Adaptive MFEA RL (AMFEARL) within the same number of training episodes. • Propose a multitask extension of ARS for training related continuous control problems. • Derived an online data-driven measurement to preempt the transfer of solutions incomplementary. • Conduct experiments on synthetic benchmarks and in the multitask Mujoco environments. [ABSTRACT FROM AUTHOR]

This paper develops a new method for lifelong learning referred to as Sparsify Dynamically Expandable Network (SDEN) via Variational Dropout, which explores a sparse model while preserving the performance. Dynamically Expandable Network (DEN) can learn a sequence of tasks via performing network retraining, network expansion by adding only the necessary neurons, and network split to effectively prevent semantic drift in an online manner. To overcome point estimation of parameters in DEN, Bayesian Compression for DEN is developed under the Bayesian framework. However, this method demands more time for model training and testing. To improve the model efficiency, we propose SDEN under the efficient sparse learning framework. We validate our SDEN in the lifelong learning scenarios with multiple frequently used benchmarks, on which it can obtain comparable classification accuracy, and less training and testing time compared with the comparison methods. Furthermore, our method can also learn a more sparse network structure which means fewer network parameters. • Comprehensively consider model performance, parameter storage space, and model training and test time. • Dynamically increase network capabilities with necessary number of neurons via Variational Dropout. • Achieve similar performance with comparative lifelong learning algorithms. • Obtain almost the minimum training and test time while considering the parameter quantity. • Learn the probability of removing a parameter or neuron from the dataset. [ABSTRACT FROM AUTHOR]

This paper presents a novel optimal reference tracking control approach resulted from the combination of a popular policy gradient Reinforcement Learning (RL) algorithm, namely Proximal Policy Optimization (PPO), and a metaheuristic Slime Mould Algorithm (SMA). One of the most important parameters in the PPO-based RL process is the learning rate, which has a big impact on how the parameters of the actor neural network (NN) are iteratively updated. In every episode of the RL process, the weights and the biases of the actor NN are multiplied with the learning rate, determining how much the learning agent will step into a certain direction computed based on previous experiences. The classical PPO algorithm usually relies on fixed values for the learning rates which rarely change, or not at all, during the learning process. However, its main drawback is that the learning agent cannot take advantage of positive momentum in the learning process by accelerating towards good learning experiences or slow down and quickly change the direction in the case of consecutive negative learning experiences. The main objective of the combination proposed in this paper is to create an adaptive SMA-based PPO approach applied to control systems, which instead of using fixed learning rate values, it uses the SMA to compute optimal values of the learning rates in each time step of the learning process based on the progress of the learning agent. This paper investigates if the adaptive SMA-based PPO control approach can be considered as an alternative to the classical PPO version, which employs fixed values of the learning rate. A comparison is carried out using control system performance indices gathered while performing an optimal reference tracking control task on tower crane system laboratory equipment. [Display omitted] • A combination of Proximal Policy Optimization and metaheuristic SMA is given. • The adaptive approach mitigates the drawbacks of constant learning rates. • SMA computes optimal values of learning rates in each learning process step. • A comparison with classical Proximal Policy Optimization is carried out. • The validation is done on real-time nonlinear tower crane position control. [ABSTRACT FROM AUTHOR]

The limited energy source used in wireless sensor networks (WSNs) is a crucial aspect when designing a routing protocol. Developing a routing protocol that tries to maximize energy utilization can significantly improve the system's lifetime and balance energy consumption. A novel geometrical-based energy-efficient routing protocol called "EVRP" is presented in this paper. The Cluster generation process is based on the k-means algorithm for nodes grouping. To choose a cluster head (CH) for each cluster, a cost function is calculated for each node considering factors such as remaining energy level for the node, and the mean distance between the node and its neighbors. In the inter-cluster routing phase, a novel geometrical-based routing protocol is introduced considering factors such as the distance to a base station (BS), remaining energy level, and distance between CHs to find the optimal path between BS and CH. The cluster structure is integrated with direct communication in addition to a dynamic clustering mechanism based on the optimum number of CHs and a load balancing mechanism to reduce the load on CHs near BS. The simulation results demonstrate that in different scenarios, the proposed protocol can significantly increase the network lifetime, network throughput, and network stability by 52 %, 45 %, and 30 % respectively compared with the EECRAIFA protocol and by 100 %, 97.6 %, 64.3 % respectively compared with D2CRP protocol. [ABSTRACT FROM AUTHOR]

As a data-centric network, the Mobile Crowd Sensing (MCS) collects and uploads sensing data through intelligent terminal devices carried by workers. However, due to resource limitations, the confidentiality, integrity and communication cost issues of sensing data have not been well coordinated and resolved in the actual MCS data collection process. In this regard, this paper proposes an edge computing-assisted MCS Chaotic Compressed Sensing Secure Data Collection scheme (CCS-SDC), which supports the secure collection of sensing data and saves communication cost. In CCS-SDC, workers first use the encryption algorithm based on chaos theory to encrypt the collected sensing data, and then adopt the hash location algorithm based on chaos theory to calculate the corresponding hash verification code of the sensing data. After receiving the encrypted sensing data transmitted by the worker, the edge server recomputes the hash verification code of the encrypted sensing data and verifies the integrity of the data, which can locate the changed sensing task data to a certain extent. Then the sensing data is compressed and sampled based on the generated chaos measurement matrix to reduce the amount of data transmission and further enhance the confidentiality of the sensing data. In addition, the same hash positioning algorithm is used between the edge server and the sensing platform to protect data integrity. For the changed data located by integrity verification, in addition to choosing to let workers re-sense and submit, the sensing platform can also choose to discard the changed sensing data under appropriate circumstances, and still reconstruct and decrypt the remaining data through the proposed algorithm to obtain effective original sensing data. The experimental evaluation results on real data sets show that CCS-SDC achieves the best effects, not only achieving lower sensing data communication cost than other related schemes, but also better protecting the confidentiality and integrity of sensing data, which is very useful for resource-constrained MCS data collection scenarios. [ABSTRACT FROM AUTHOR]

In this paper, we assume that a team of drones equipped with sensing and networking capabilities explore an unknown area via onboard sensors for surveillance, monitoring, target search or data collection purposes and deliver the sensed data to a ground control station (GCS) over multi-hop links. We propose a multi-drone path planner that jointly optimizes area coverage time and connectivity among the drones. We propose a novel connectivity metric that includes not only percentage connectivity of the drones to GCS, but also the maximum duration of consecutive time that the drones are disconnected from the GCS. To solve this optimization formulation, we propose a multi-objective evolutionary algorithm with novel operations. We use our solver to test single, two and many objective path planning problems and compare our Pareto-optimal solutions to benchmark weighted-sum based solutions. We show that as opposed to the single solution that weighted-sum methods provide based on prior information from the user, the proposed evolutionary multi-objective optimizers can provide a diverse set of solutions that cover a range of mission time and connectivity performance illustrating the trade-off between these conflicting objectives. The end-user can then choose the best path solution based on the mission priorities during operation. [ABSTRACT FROM AUTHOR]

The data security of CAN bus system is receiving increasing attention with the rapid development of Internet of Vehicles (IoV). However, traditional ciphers are not the best choice due to the limitations of computation, real-time, and resources of Electronic Control Units in vehicles. Thus, this paper proposes a lightweight block cipher IoVCipher to protect the security of IoV. It is designed focus on the latency and area in round-based architectures (both encryption and decryption) to meet this resource-constrained environments. For this purpose, two S-boxes with low latency and tiny area are constructed in this paper, one involution and one non-involution. Considering the decryption latency, a low latency subkey generation method is designed. In addition, this paper proposes a new extended MISTY structure that makes the encryption and decryption of hardware implementations similar. In comparison to other low-latency lightweight block ciphers such as PRINCE, QARMA, MANTIS and LLLWBC, IoVCipher achieves an effective balance between latency and area in the round-based architecture, and IoVCipher has low latency, low area, and low energy in the fully unrolled architecture. Finally, IoVCipher is implemented on a real-time speed acquisition and encryption testbed to simulate encrypted transmission of real-time speed in a CAN bus environment. [ABSTRACT FROM AUTHOR]

Due to the wide range of intensive computational applications and ubiquitous connectivity of the Internet of Things (IoT) paradigms, it has become crucial to develop a new platform that can achieve low delay, high network throughput, and enhanced quality of service (QoS). This paper proposes a joint user association, service caching, and task offloading strategy to reduce delay and enhance users' QoS in multi-tier communication and multi-tier edge computing heterogeneous networks (HetNets). The considered system model consists of multi-users with different tasks and service data sizes communicating in a heterogeneous network of one massive multiple-input multiple-output (M-MIMO) macro base station and some small base stations. The proposed work investigates user association, power allocation, optimum service data caching, and task offloading strategies at the computing network edges. Thereby, the objectives of this work are to propose an efficient framework to reduce the system delay, increase the network throughput, and meet the user requirements in multi-tier communication and multi-tier edge computing heterogeneous networks. The simulation results show that the proposed algorithm outperforms the state-of-the-art with a 49.48% decrease in system delay, 80% reduction in cost and hardware complexity in terms of the reduced number of installed antennas, and 48.58% enhancement in the network throughput. [ABSTRACT FROM AUTHOR]

Energy efficiency (EE) in large-scale high-density ad hoc networks is an NP-hard problem of fractional function optimization. Most traditional algorithms in the literature focus only on power control, mainly optimizing the solution of power control with the objective of sum-rate maximization or weighted sum-rate maximization, etc., and thus are unsuitable for solving fractional function EE optimization problems. In this paper, the architectural design of the neural network algorithm is directly guided with the objective of maximizing EE. Specifically, we first prove that the EE problem can be described as a graph optimization problem with permutation equivariance. Then, we propose an energy-efficient optimized graph convolutional neural network (EEOGCN) algorithm that satisfies permutation equivariance. Experiments show that the algorithm has better energy efficiency performance compared to other algorithms, can be directly applied to weighted energy efficiency optimization and beamforming energy efficiency optimization problems, and exhibits good generalization performance and scalability to large-scale high-density problems. In particular, the proposed method is computationally efficient and can solve the weighted energy efficiency optimization problem for 3000 transceiver pairs on a single GPU within 2 ms. [ABSTRACT FROM AUTHOR]

The number of smart wireless devices (WDs) has enormously increased over the last few years due to the advancement of 5G/B5G networks. The advanced applications of such smart WDs, e.g., augmented reality, virtual reality, online gaming, etc., demand excessive resources. Although the WDs are equipped with limited resources, the evolution of edge computing and offloading techniques enables the WDs to offload their resource-intensive tasks to the nearby edge node. These edge nodes might experience higher loads and delays when WDs generate a huge number of tasks. Moreover, the wireless channel bandwidth and transmission data rate of the wireless channels are also limited. Therefore, optimizing the use of available bandwidth as a valuable resource and reducing latency emerge as crucial objectives while offloading tasks. In this paper, a delay-aware resource-constrained offloading problem for edge–cloud systems is mathematically formulated as a 0–1 integer linear programming, and it is shown to be NP-complete. Then, a delay-aware resource-constrained offloading algorithm based on a discretized differential evolution (DARC-DE) is designed. The objectives of the DARC-DE are to maximize the utilization of the resources as bandwidth and minimize the delay. The vectors are efficiently encoded along with the decoding technique. The fitness function is designed by considering execution, offloading, queuing, transmission delay, and bandwidth utilization. The DARC-DE is shown to be executed in polynomial time. To evaluate DARC-DE, extensive simulation is performed in two different scenarios with varying numbers of tasks and edges. Simulation results demonstrate that the proposed DARC-DE can minimize total delay by 15% to 40% in comparison to particle swarm optimization, genetic algorithm, and bees algorithm, respectively. Simulation results also indicate a significant improvement in bandwidth utilization. Taguchi method and alternative average convergence rate are conducted. The statistical tests—analysis of variance, post-hoc, and Friedman tests—are also performed. • DARC-DE is designed for efficient offloading in edge–cloud systems. • Efficient encoding and decoding of discretized solution vectors are given. • The wireless channel bandwidth as a resource is dynamically allocated. • Execution, offloading, queuing, and transmission delay are considered. • Derivation of a novel fitness function that considers delay and bandwidth as resources. [ABSTRACT FROM AUTHOR]

Lately, the continuous temporal data from motion sensors in wearable devices has been great interest for the research community due to its demand for analyzing human activities in several applications e.g. healthcare, sports, and surveillance. Numerous solo deep learning models have been proposed in the literature to extract an adequate feature representation from temporal sensory data, however, they are limited to encode only one aspect of the data, and not sufficient to capture the in-depth association between the patterns. This paper presents an ensemble deep model to encode the complex patterns and dependencies in temporal data. An ensemble model refers to the merging or blending of separate deep models, aiming to utilize their unique strengths and abilities to construct a more resilient and effective solution. The contributions of this paper are two folds: First, a systematic evaluation of five different ensemble models is presented to encode the raw sensory data. Secondly, this study proposed a lightweight, multi-layer, and hybrid LSTM-GRU model to recognize the human activities using time series sensory data. Specifically, the raw sensory data is fed into a two-layer LSTM followed by GRU layers. To avoid over-fitting and accelerate convergence, we employed dropout and batch normalization layers. An extensive evaluation of all the five ensemble models is carried out on two large benchmark datasets, namely: UCI-HAR and WISDM. The proposed LSTM-GRU architecture achieved the recognition accuracy of 99.06% and 96.61% on the WISDM and UCI-HAR, respectively. The performance comparison with existing state-of-the-art techniques confirms the superiority and efficacy of the proposed algorithm. • A comprehensive literature review of existing HAR techniques is conducted. • A systematic evaluation of five different ensemble models is presented. • The optimal network configuration is investigated to improve accuracy and efficiency. • The performance of proposed model is evaluated on two large and diverse datasets. • A comparison of various architectures is presented to analyze the sensory data. [ABSTRACT FROM AUTHOR]

The structural deformation of a dam directly affects its lifespan and safety, making its accurate prediction crucial. Traditional prediction methods often overlook the nonlinearity and non-smoothness of deformation data. Moreover, the irregular intervals within the historical deformation data used for model training can reduce prediction accuracy. To address these issues, we propose a hybrid deep learning model that uses signal decomposition and reconstruction to enhance dam deformation prediction. This model employs a long short-term memory (LSTM) neural network optimized using a complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN). The self-attention mechanism in the LSTM model effectively captures the temporal features of dam deformation, alleviating the difficulties generated by the irregular intervals within the historical data used in model training. Furthermore, considering the lag effect of influencing factors on dam deformation and the differences among various measurement points, we propose a CEEMDAN-based feature selection method. Using 13 years worth of data from the Shuibuya Dam, we evaluate the accuracy and effectiveness of the CEEMDAN-Self-attention-LSTM model using indicators, such as MAE, RMSE, MAPE, and R 2 , and compared it with existing models. The experimental results show that this model reduces prediction error by more than 53.62%. • This paper proposes a dam deformation prediction model based on the CEEMDAN-Self-attention-LSTM model. • The CEEMDAN algorithm effectively reduces the complexity of dam deformation data. • The modelis able to better capture the cyclical changes between time steps. • A feature selection method based on CEEMDAN fully considers the lagging impact of environmental factors on dam deformation. • The predictive capability of this method is analyzed with a case study of the Shuibuya concretefaced rockfill dam. [ABSTRACT FROM AUTHOR]

Attribute reduction is a hot research topic in data mining, in which rough set theory-based attribute reduction methods have been widely focused. The neighborhood rough set (NRS) model has good generalization performance and practicality in uncertainty reasoning, so it is often used for attribute reduction in recent years. Calculating the distance between samples in different attribute spaces is a key step in the NRS-based attribute reduction methods, which directly affects the performance of the reduction algorithm. However, the NRS model uses a fixed computational paradigm in calculating the distance between samples and does not consider the influence of labels in the attribute spaces on the distance calculation, which is not conducive to improving the performance of the reduction algorithm. Distance metric learning takes full account of the labels information in the multi-dimensional attribute space, and it can learn the distance between samples by taking into account the integrated principle that samples with the same label are closer and samples with different labels are further away, which helps to reduce the classification uncertainty. Inspired by this, this paper firstly incorporates distance metric learning into the NRS model from the perspective of multi-dimensional attribute space and proposes a distance metric learning-based multi-granularity neighborhood rough set (DmlMNRS) model. The related properties of the DmlMNRS model are also introduced and proved. Then, the DmlMNRS-based attribute reduction criterion and the significance of the attributes are defined. A DmlMNRS-based heuristic attribute reduction (DMNHAR) algorithm is designed based on this. Finally, experiments are performed on fifteen publicly datasets, and the experimental results show that the proposed algorithm has better robustness and classification performance. • A neighborhood granulation method using Distance Metric Learning (DML) is proposed. • A novel DML-based multi-granularity neighborhood rough set model is proposed. • A multi-granularity heuristic attribute reduction algorithm based on DML is designed. • Numerical experiments show that our algorithm have good classification performance. [ABSTRACT FROM AUTHOR]

Oxygen excess ratio (OER) is regarded as a significant indicator of the safety and efficient operation of the open-cathode proton exchange membrane fuel cell (PEMFC) system. This is because inappropriate OER is prone to the phenomenon of oxygen starvation and oxygen saturation under load conditions change frequently. To avoid the phenomenon, this paper integrates linear active disturbance rejection control (LADRC) and proportional integral differential (PID) control to regulate OER of fuel cell air supply system, which effectively combines model-based and data-driven LADRC structure. As the controller parameters directly affect the control performance and it is difficult to obtain the optimal control parameters. Especially, when the load conditions change, the control parameters are difficult to adapt the load changes. To this end, a snake optimization (SO) algorithm with easy implementation and fast convergence is used to realtime update the controller parameters, which can further improve the control effects. Firstly, an equivalent linear model of fuel cell air supply system is established. Secondly, a hybrid LADRC method is developed by combining the merits of the LADRC and PID methods. Thirdly, the SO algorithm is used to optimize the controller parameters, which can achieve the optimal control parameters. Finally, the results revealed that the proposed method has a faster response with less noise sensitivity against the load changes and stronger robustness than the other methods. • A fusion control strategy is proposed to regulate fuel cell air supply system. • The parameters of the controller are optimized by snake optimization algorithm. • The proposed method can improve the transient dynamic response of OER control. [ABSTRACT FROM AUTHOR]

Multivariate time series usually have entangled temporal patterns and various anomaly types. Meanwhile, they often contain both continuous and discrete features. Many existing methods directly model correlations in complex multivariate time series to conduct anomaly detection. Decomposing time series into different components, such as the overall trend and fluctuations, can contribute to better extracting semantic information and detecting anomalies. Existing decomposition-based anomaly detection methods still have several limitations. First, they directly decompose all features without considering that discrete features are unsuitable for decomposition because they do not have trends or fluctuations. Second, they adopt the same networks for different components with different characteristics, limiting their ability to extract semantic information. Moreover, due to the nature of Transformers, existing reconstruction-based methods using Transformers rarely form information bottlenecks, reducing the differentiation between the reconstruction errors of normal data and anomalies. This paper proposes a multivariate time series anomaly detection method with separation, decomposition, and dual Transformer-based autoencoder (SDDformer). Different from existing methods, SDDformer separates continuous and discrete features and only decomposes continuous features into trend and residual components. Considering the different characteristics of different components, SDDformer adopts Crossformer and the vanilla Transformer as the backbone of two different autoencoders to reconstruct the trend and residual components. Information bottlenecks are better formed using an extra token as the latent variable between the encoder and the decoder. SDDformer regards reconstructing a discrete feature as a classification task and calculates Cross-Entropy as its reconstruction error. Three different metrics are adopted in this paper to compare SDDformer with a variety of typical anomaly detection methods on public data sets, and the experimental results prove that SDDformer can achieve state-of-the-art performance. • A two-stream multivariate time series anomaly detection method is proposed. • Continuous and discrete features are separated before decomposition. • Only continuous features are decomposed into the trend and residual components. • Two different autoencoders are proposed to reconstruct different components. • The reconstruction of a discrete feature is regarded as a classification task. [ABSTRACT FROM AUTHOR]

Multilabel classification has recently attracted great attention from the data mining research community. Multilabel classification is concerned with learning where each instance can be associated with multiple classes (or labels). Class-imbalance problems appear in any classification task when the class distribution of the instances is very different. In multilabel classification, this problem is ubiquitous, as a large percentage of labels suffer from a class-imbalanced distribution. The adaptation of single-label methods to deal with the class-imbalance problem in multilabel learning is problematic as many of their basic concepts are not easily transferred. In this paper, we propose the use of evolutionary computation to simultaneously oversample the minority class and undersample the majority class for multilabel problems. Letting the algorithm autonomously select the instances to undersample and oversample allows us to extend these two successful paradigms to the multilabel task. An extensive comparison setup of 35 datasets shows the advantages of using this approach to deal with class-imbalance datasets for multilabel problems compared with previously published methods as well as the basic classification algorithms with the original datasets. • We propose a new approach for dealing with the class-imbalance problem in multi-label datasets. • The method is based on an evolutionary simultaneous under and oversampling. • The method is scalable and achieves better results than current approaches. [ABSTRACT FROM AUTHOR]

In recent years, emotion recognition based on electroencephalogram (EEG) has become an important research field. This paper proposes an innovative multi-scale emotion recognition method (MS-ERM), which is based on a deep learning model. First, we divide the EEG signal into time windows of 0.5 s in different frequency bands to extract the differential entropy feature and embed the feature into the brain electrode map to express spatial information. Then, the features of each segment are used as input to the new deep learning model (MS-TimesNet). The model combines multi-scale convolution and TimesNet network to effectively extract dynamic time features, cross-channel spatial features, and complex time features in 2D space. Through extensive tests on the DEAP dataset, we prove that this method is superior to existing methods in terms of sentiment classification performance. In the arousal and valence classification, the average classification accuracy of subject-dependent tests reached 91.31% and 90.45%, respectively, while in subject-independent tests, the average classification accuracy was 86.66% and 85.40%, respectively. Code is available at this repository: https://github.com/hyao0827/MS-ERM. • An emotion recognition method MS-ERM based on deep learning is proposed. In this method, differential entropy (DE) is embedded in the electrode map to fuse spatial information as the input of the MS-TimesNet model. • The MS-TimesNet model is proposed, which combines multi-scale convolution and TimesNet network to extract complex spatio-temporal features at different levels. • Extensive tests on the DEAP dataset verify the superiority of MS-ERM, which is significantly higher than the existing methods. [ABSTRACT FROM AUTHOR]

There is often a lack of explanation when artificial intelligence (AI) is used to diagnose skin lesions, which makes the physician unable to interpret and validate the output; thus, diagnostic systems become significantly less safe. In this paper, we proposed a deep inherent learning method to classify seven types of skin lesions. The proposed deep inherent learning was validated using different explanation techniques. Explainable AI (X-AI) was used to explain decision-making processes at the local and global levels. In addition, we provide visual information to help physicians trust the proposed method. The challenging dataset, HAM10000, was used to evaluate the proposed method. Medical practitioners can better understand the mechanisms of black-box AI models using our simple, stage-based X-AI framework. They can trust the proposed method because the rationale for its decisions is explained. • A novel inherent learning technique with multiple convolution filters is proposed. • The proposed algorithm overcomes image shortages and degradation. • Different explanations prove the proposed model's capability to detect lesion areas in the skin accurately. • The proposed deep inherent learning method is less error-prone than shallow networks. • The proposed method's ability to classify skin lesions accurately is proven. [ABSTRACT FROM AUTHOR]

Knowledge distillation is one of the most persuasive approaches to model compression that transfers the representational expertise from large deep-learning teacher models to a small student network. Although numerous techniques have been proposed to improve teacher representations at the logits level, no study has examined the weaknesses in the representations of the teacher at the feature level during distillation. On the other hand, in a trained deep-learning model, all the kernels are not uniformly activated to make specific predictions. Transferring this knowledge may result in a student learning a suboptimal intrinsic distribution and restrict the existing distillation methods from exploiting their highest potential. Motivated by these issues, this study analyses the generalization capability of teachers with or without a uniformly activated channel distribution. Preliminary investigations and theoretical analyses show that partly uniforming or smoothing feature maps offer improved representation that enriches the generalization capability. Based on these observations, it is hypothesized that distillation-based explicit supervision using smoothed feature maps and cross-entropy loss plays a significant role in improving generalization. Hence, this paper proposes a novel technique called P artly U nified R ecalibrated F eature (PURF) map distillation. The proposed method recalibrates the feature maps by intercommunicating the representational cues among nearest-neighbor channels. PURF increases the performance of state-of-the-art knowledge distillation (KD) methods across architectures by improving generalization, model compression, few-shot training, transferability, and robustness transfer on standard benchmark datasets. PURF achieves 1.51% average accuracy improvements on seven diverse architectures in image classification tasks. PURF increases the performance of the state-of-the-art knowledge distillation methods by an average accuracy of 1.91% across architectures. Moreover, PURF achieves an average of 2.02%, and 0.96% higher accuracy in transferability and robustness tasks, respectively, on standard benchmark datasets. • A novel technique, PURF, helps SOTA-KD methods maximize their potential. • Generalization technique without decreasing training performance. • Improves the performance of knowledge distillation tasks. • A generalization method by imposing explicit feature-level smoothness constraints. • An extensive analysis of feature-level smoothness as a regularization technique. • Can be employed as the top-head unit for the existing distillation techniques. [ABSTRACT FROM AUTHOR]

The scope of this paper is to test the efficiency of three different neural networks in the area of cloud-to-ground/intra-cloud lightning classification using a not very large data set of electric field data. Effective neural network learning verified for a small lightning event dataset but with whole 2-seconds particular record length is an important feature of fast and efficient procedures implemented into lightning location systems among which discrimination between cloud-to-ground and intra-cloud lightning is one of the most important task. Measurement data obtained at the Lightning Observatory in Rzeszow were used for the analysis. About 100 extremely low/medium frequency bandwidth electric field of cloud-to ground lightning registrations and the same number of intra-cloud lightning events were taken into account. During the study 81% of those registrations were dedicated for learning of neural networks while the remaining 19% were put into the testing set. All data were preselected manually with consideration of the electric field variation and information from the lightning location system database. Lightning events were selected uniformly with the well-pronounced presence of their basic components and reported distance to the registration station which ensured a wide variation of expected lightning electric field signatures. The multilayer perceptron neural network, the radial basis function neural network and the convolutional neural network were tested and compared in different configurations while distinguishing lightning. The results were compared with other conventional classification approaches, such as traditional machine learning methods as well as one a little more contemporary architecture, the long short-term memory neural network. The multilayer perceptron neural network achieved the best detection accuracy overall. Presented research is an extension of lightning identification studies available so far mainly based on the convolutional neural network by the results achieved using the multilayer perceptron neural network, the radial basis function neural network, the long short-term memory neural network and several machine learning-based classification methods. • Research based on electric field data for CG and IC lightning from Rzeszow. • CG/IC lightning discrimination was verified in presence of a small dataset. • MLP, RBF, LSTM and convolutional neural networks as well as ML were verified. • 3-layer MLP neural networks showed the best performance overall. • Analysis results can be adapted for a real-time lightning location system purposes. [ABSTRACT FROM AUTHOR]

In the semiconductor manufacturing process, various kinds of metrology and test data can form different types of wafer maps. By analyzing the correlation of multiple types of wafer maps, especially for the wafer bin maps (WBMs) and the wafer profile maps, the faulty inline manufacturing steps strongly correlated with end-of-line yield can be found for yield improvement. This paper proposes a correlation analysis method based on a deep reconstruction model to integrate the knowledge among multiple types of wafer maps for root cause analysis. First, the sparsely sampled wafer profile maps are restored to original wafer profile maps through the deep reconstruction model for obtaining more information. And then, the correlation between the WBMs and the reconstructed profile maps is calculated. The process step whose wafer profile maps have the highest correlation with WBMs is highly related to the fault. Experiments on the real-world dataset demonstrate that the proposed method can restore the wafer profile map well and has high accuracy in matching the relevant process wafer profile map based on the WBMs. • The deep reconstruction model reconstructs full wafer profile maps. • The correlation analysis framework correlates wafer maps from different domains. • The deep reconstruction model's effectiveness is proven compared to existing methods. [ABSTRACT FROM AUTHOR]

The current deep learning-based intelligent diagnosis algorithms depend on large amounts of well-labeled data, but they may not perform well in engineering practice where the fault data available from a single device is limited. The significant difference in data distribution between different devices makes it more difficult to transfer diagnosis across devices. Therefore, aiming at the problem that the great difference in the data distribution of source and target domains in cross-device transfer diagnosis leads to the low diagnosis accuracy, this paper investigates a deep transfer network model based on multi-domain information fusion and multi-kernel maximum mean discrepancy (MK-MMD) for fault diagnosis of rolling bearings across different devices. Firstly, to address the problem that single-domain information is difficult to adequately characterize rolling bearing health status information of different devices, a multi-domain information feature extractor based on multi-attention mechanism is constructed to capture the important features of vibration signals in different transform domains. Secondly, the extracted multi-domain features are input into the bidirectional gated recurrent unit (BiGRU) network for feature fusion, and MK-MMD is used to narrow the distribution distance between the source domain and target domain to obtain domain invariant features, and then achieve the transfer diagnosis of rolling bearing faults between different devices. Finally, the method investigated is verified by rolling bearing fault transfer diagnosis tests between different devices, and the results show that the suggested method can adapt to the feature distribution between different domains, and improve the transfer diagnosis accuracy between different devices. • A cross-device fault transfer diagnosis method for rolling bearings is proposed. • A multi-domain feature extractor is built to fully mine key fault information. • A multi-domain feature deep fusion strategy based on BiGRU is designed. • MK-MMD is adopted to obtain domain invariant features of source and target domains. • The cross-device transfer diagnosis tasks verify the superiority of the method. [ABSTRACT FROM AUTHOR]

In real production, assembly and machine deterioration are extremely common phenomenon and flexible flow shop scheduling problem (FFSP) is also extensively investigated. However, assembly FFSP with machine deterioration effect (AFFSP-DE) is neglected. This paper addresses an AFFSP-DE with makespan criterion and proposes a discrete artificial bee colony (DABC) algorithm. In DABC, the number of employed (onlooker) bees is dynamically changeable according to evolutionary quality to adaptively allocate computing resources for two kinds of bees, and a new scout bee phase is designed to improve convergence speed. Finally, the relationships between makespan and different deterioration rates are analyzed systematically by extensive experiments, and the results demonstrate that DABC outperforms the existing algorithms over 82% test instances in solving the AFFSP-DE. • We investigate an assembly flexible flow shop scheduling problem with deterioration effect. • A discrete artificial bee colony algorithm is proposed to solve the considered problem. • The employed and onlooker bee phases are designed to allocate computing resource. • A scout bee phase is given to prevent search stagnation of proposed algorithm. • Relationships between makespan and deterioration rates are analyzed systematically. [ABSTRACT FROM AUTHOR]

Accurate delay evolution prediction plays a pivotal role in train rescheduling decision-making for the railway network. Existing studies on delay prediction predominantly centered around predicting delays for each train in the subsequent stations (i.e., following a train-oriented perspective). Furthermore, train operations in the railway network involve different types of entities (stations, trains, etc.), making the current graph/network models with homogenous nodes (i.e., the same kind of nodes) incapable of effectively capturing the interactions between the entities. This paper develops a network-oriented model to investigate the train delay evolution on railway networks, by predicting the delays of running trains in the network after a given time interval. The proposed model combines the GraphSAGE graph neural network (GNN) and the heterogeneous graph neural network (HetGNN) architecture, thus called SAGE-Het, enabling it to capture interactions between heterogeneous nodes (i.e., different types of nodes) based on different edges (e.g., edges between trains, trains, and stations). Additionally, SAGE-Het allows for flexible inputs in contrast to conventional machine learning techniques, whose inputs must meet the consistent dimension requirement (e.g., in the form of rectangular or grid-like arrays). The performance and robustness of the suggested SAGE-Het model are assessed in experiments on the data from two sub-networks of the China railway network. The experimental results demonstrate that SAGE-Het outperforms the existing delay prediction methods and some advanced HetGNNs used for prediction tasks in other domains; SAGE-Het demonstrates excellent scalability, capable of handling various types of nodes; the predictive performances of SAGE-Het under different prediction time horizons (10/20/30 min ahead) all exhibit better performance over other baselines; the accuracies are over 90 % under the permissible 3-minute errors for the three prediction time horizons. Specifically, the impact of train interactions on delay evolution is investigated based on the flexible input characteristic of the proposed model. The results illustrate that train interactions become subtle with the increase of train headways. This finding directly contributes to decision-making in situations where conflict resolution or train-canceling actions are needed. • A novel network-oriented approach is proposed to investigate the delay evolution in the railway network. • The proposed heterogeneous graph neural network approach allows for diverse heterogeneous nodes and flexible inputs. • The proposed approach has better predictive performance than the commonly used delay prediction methods. • The importance of train interaction on delay evolution is explored based on the proposed approach. [ABSTRACT FROM AUTHOR]

Effective vehicle detection is extremely vital in cities. However, existing vehicle recognition models exhibit limited functionality and have not been effectively integrated with language models to aid in vehicle monitoring within cities. To address these challenges, this paper proposes a semi-supervised mixture model with detection, segmentation, and question-answering capabilities, referred to as the Multi-task Visual Language Model (MTVLM), for vehicle recognition. Further, a vehicle dataset containing 8425 vehicle images from 13 different categories is introduced, which is collected across various urban scenarios and weather conditions. MTVLM consists of a detector, a language model, and a segmentation model. The detector YOLODS is based on an improved You Only Look Once algorithm (YOLOv4), which has a lower number of parameters and higher detection accuracy. The detection results of YOLODS are then used as prompts for the pre-trained language model, a distilled version of Bidirectional Encoder Representations from Transformers (DistilBERT), to answer questions related to the detected vehicles. Furthermore, the coordinates of bounding boxes obtained by YOLODS are input into the Segment Anything Model (SAM) as prompts to segment vehicles from images. Experimental results show that MTVLM achieves a mean average precision (mAP) of 97.12 % on the vehicle dataset, surpassing other state-of-the-art (SOTA) detectors. The high-precision detector also ensures highly accurate text prompts for input to DistilBERT and SAM. By employing text prompts, the potential and performance of the pre-trained models DistilBERT and SAM can be fully explored and utilized to better complete question-answering and segmentation tasks. • A visual language model MTVLM is created to identify and monitor urban vehicles. • For accurate vehicle detection, a high-precision object detector is constructed. • A dataset for vehicle recognition is constructed from 13 different vehicle categories. [ABSTRACT FROM AUTHOR]

Nonlinear spiking neural P (NSNP) systems are a novel class of neural-like computing models that exhibit nonlinear dynamic behaviours. Further, reservoir computing (RC) is a new type of recurrent neural network (RNN) whose dynamic behaviour is provided by a reservoir of neurones. An interesting question is whether NSNP systems can be embedded into an RC framework to develop an efficient prediction model for time series forecasting problems. To address this problem, this paper first defines a variant of NSNP systems with two outputs, called NSNP-TO systems, in which a nonlinear spiking mechanism with two outputs is the key feature. Based on NSNP-TO systems, a new variant of RC models is proposed based on NSNP systems, called the RC-NSNP model. Owing to the use of spiking neurones, the RC-NSNP model comprises state-updating and output equations that are different from those of the existing RC models. Using a similar structure, the proposed RC-NSNP model can be implemented in the RC framework, and a similar training method can be utilised to train it. Time series forecasting tasks are performed to verify the effectiveness of the proposed RC-NSNP model. The proposed RC-NSNP model achieved the best performance on nine of the twelve time series problems and obtained the second-best performance on the other two time series. • Nonlinear spiking neural P systems with two outputs are defined, called NSNP-TO systems. • NSNP-TO systems can be used as the reservoir of RC model. • A new RC model is proposed for time series forecasting, called RC-NSNP model. [ABSTRACT FROM AUTHOR]

This paper addresses a two-dimensional rectangular cutting problem with guillotine constraint, applied to the transformer industry. The problem requires a set of rectangular items to be cut from larger and variable-sized rectangles, known as bins. The goal is to minimize the number of bins used to cut all required items. An effective discrete artificial bee colony is proposed to solve the problem. The approach uses a decimal encoding method where each integer represents a type of bin or item, and decodes the list of integers by a constructive heuristic to generate cutting patterns with guillotine-cut constraint. In addition, a copy strategy is designed to improve the solution quality. The approach's performances and the effectiveness of the strategy are evaluated by three datasets generated from real-world data. The experimental results show that the discrete artificial bee colony is superior to other competitive algorithms, and that the copy strategy can produce higher quality solutions. • A cutting problem with guillotine arisen in reality is studied. • A creative heuristic algorithm is proposed to generate cutting pattern. • A discrete artificial bee colony is proposed to solve the problem. • The performance of approach is assessed by datasets obtained from factory. [ABSTRACT FROM AUTHOR]

Density-based clustering is a widely explored domain in the field of machine learning, with numerous methods proposed to address challenges in clustering under varying density conditions. These methods primarily categorize samples based on their local density. However, challenges arise when dealing with boundary samples. The density of certain boundary samples is typically lower, which can lead to misclassification as noise. Moreover, in situations where two clusters have similar densities and are in close proximity, accurately classifying the boundary points between them becomes a challenging task. In this paper, we introduce an enhanced approach based on k -nearest neighbors to address this challenge within density-based clustering. In our method, upon the formation of a new cluster, we identify boundary samples by examining the spatial relationships between each sample and its k -nearest neighbors, as well as their connections to the newly established cluster. Once all clusters are formed, we refine the classification of these boundary samples by adjusting or retaining their assigned labels based on their k -nearest neighbors. Experimental evaluations on synthetic and real-world datasets demonstrate the effectiveness of our proposed method. • A new density clustering method based on boundary samples verification is proposed. • The method can handle datasets with different densities and shapes. • The method only requires one integer parameter, easy to use. • Outperforms existing methods on various benchmark datasets. [ABSTRACT FROM AUTHOR]

The computational analysis of real-world engineering problems often relies on time-consuming simulation calculations. This presents challenges in balancing computational burden and precision when applying traditional metaheuristic algorithms to large-scale complex problems. While there have been numerous surrogate-assisted metaheuristic optimization algorithms proposed, most of them are designed for solving expensive problems within 30–50 dimensions. In this paper, a novel approach called Parallel-multi-swarm Gaussian Process Regression surrogate-based particle swarm optimization (Parallel-MS-GPRS-PSO) is proposed for high-dimensional problems. This approach combines a standard particle swarm optimization (PSO) algorithm, multi-swarm cooperative PSO (MSCPSO), and a GPR surrogate-based PSO (PSO-GPR). By integrating PSO-GPR and MSCPSO, the proposed method aims to effectively explore and exploit the search space while enhancing the global and local performance of the surrogate model. To validate the effectiveness of the proposed algorithm, it is compared with several existing PSO algorithms on seven benchmark functions and an engineering problem. The results illustrate that the Parallel-MS-GPRS-PSO approach outperforms the existing algorithms. Moreover, the method shows promising performance in computationally expensive engineering optimization problems with 288 variables. • Parallel-MS-GPRS-PSO approach is designed to address high-dimensional problems. • The method combines PSO, MSCPSO, and PSO-GPR to enhance surrogate model performance. • The proposed algorithm outperforms existing PSO approaches on benchmark functions. • The method shows promising performance in complex engineering optimization problem. [ABSTRACT FROM AUTHOR]

Instance selection is an important preprocessing technology in data mining and machine learning. In this paper, we proposed a novel evolutionary based instance selection algorithm for big data. First, we defined a coarse granularity chromosome structure to reduce the size of search space and costs of chromosome operations(recombination and mutation, etc.). Then a stratified evolution strategy was proposed to remove the hyper parameter in classic fitness function and achieve precise control over the reduction ratio of instances. Finally, a sampling-based fitness function was proposed to reduce the time complexity. Experimental results shown that our new algorithm is efficient to complete the instance selection task on data set with millions of instances in minutes-level. The 10-fold cross-validation also proved that the selection results on many datasets have high nearest neighbor classification accuracy. • We proposed a distributed coarse granularity chromosome structure to reduce search space and ensure load balance. • We proposed a stratified evolution strategy to transform multi-objective to multiple single-objective tasks. • We proposed a sampling-based fitness function with less time complexity [ABSTRACT FROM AUTHOR]

This paper introduces a novel meta-heuristic optimization algorithm named Clustering Wavelet Opposition-based Marine Predators Algorithm (CWOMPA) to address some limitations present in the well-established Marine Predators Algorithm (MPA). CWOMPA incorporates three key strategies: a fuzzy clustering approach for escaping local optima, using wavelet basis function-based impact coefficient adjustment for elites to prevent premature convergence, and finally an adaptive opposition-based learning strategy for maintaining population diversity. Compared with some recent meta-heuristic algorithms, extensive evaluations conducted affirm that CWOMPA achieves the best Friedman rank, 4.30 and 1.95 respectively, on 23 benchmark functions and the CEC 2017 benchmark set. Not only does CWOMPA demonstrate significant effectiveness on six constrained problems from the CEC 2020 real-world benchmarks, but also when applied to 14 medical datasets, it gains superior Friedman ranks in terms of selected features, classification accuracy, F-Score, and objective function value, 2.5, 2.25, 2.96, 2.93 respectively, outperforming other common methods. Finally, CWOMPA exhibits the highest classification accuracy across all datasets and the best F-Score performance in nine datasets compared to traditional feature selection algorithms. The obtained results reveal that CWOMPA is a powerful and versatile optimization algorithm with significant potential in various real-world applications, including feature selection for medical datasets. [Display omitted] • An improved MPA called CWOMPA is proposed to improve the optimization performance. • CWOMPA incorporates FCM, wavelet basis function, and AOBL to enhance the MPA. • CWOMPA can be an effective tool in addressing feature selection challenges. [ABSTRACT FROM AUTHOR]

Dynamic multi-objective optimization problems (DMOPs) can be utilized to model certain real-world problems that have a dynamic nature. Algorithms for solving DMOPs can be evaluated and improved by comparing their performance on different benchmarks. However, some existing benchmarks for DMOPs have the limitation of non-uniform weights for decision variables. Additionally, dynamic many-objective optimization problems (DMaOPs) involve more than three objectives, but only a few existing benchmarks can be extended to accommodate DMaOPs. Furthermore, some existing performance measures for DMOPs may not effectively compare the relative performance differences between multiple algorithms or evaluate the search uniformity among different objectives. In this paper, we propose improvements to an existing benchmark for DMOPs by expanding the impact range of decision variables. Moreover, a benchmark framework that can be extended to accommodate DMaOPs is proposed, thus addressing a research gap between the optimization of DMOPs and DMaOPs. Additionally, a set of performance measures for DMOPs are proposed, which can evaluate the relative performance and search uniformity of multi-objective optimization algorithms. By comparing the performance of state-of-the-art and commonly used algorithms on test problems, we can gain a better understanding of the characteristics and strengths and weaknesses of the algorithms and test problems. • A set of improved DMOPs test problems are proposed in this paper. • A scalable DMaOP benchmark framework is proposed in this paper. • A set of performance measures that applies to both MOPs and DMOPs are proposed. [ABSTRACT FROM AUTHOR]

Normalization of objectives plays a crucial role in evolutionary multi-objective optimization (EMO) to handle objective functions with different scales, which can be found in real-world problems. Although the effect of normalization methods on the performance of EMO algorithms has been investigated in the literature, that of preference-based EMO (PBEMO) algorithms is poorly understood. Since PBEMO aims to approximate a region of interest, its population generally does not cover the Pareto front in the objective space. This property may make normalization of objectives in PBEMO difficult. This paper investigates the effectiveness of three normalization methods in three representative PBEMO algorithms. We present a bounded archive-based method for approximating the nadir point. First, we demonstrate that the normalization methods in PBEMO perform significantly worse than that in conventional EMO in terms of approximating the ideal point, nadir point, and range of the PF. Then, we show that PBEMO requires normalization of objectives on problems with differently scaled objectives. Our results show that there is no clear "best normalization method" in PBEMO, but an external archive-based method performs relatively well. • This paper investigates the effectiveness of three normalization methods in three representative PBEMO algorithms. • We demonstrate that the normalization methods in PBEMO perform significantly worse than that in conventional EMO in terms of approximating the ideal point, nadir point, and range of the PF. scaled objectives. • We show that there is no clear "best normalization method" in PBEMO, but an external archive-based method performs relatively well. [ABSTRACT FROM AUTHOR]

Three-way decision, as an extension of traditional two-way decision, was proposed by Yao in 2009, which can effectively avoid unnecessary losses caused by incorrect decisions in the decision-making process. Meanwhile, incomplete hybrid information systems represent the database of the relationship between objects and attributes, which refers to a system with multiple data and missing data. In this paper, based on the fact that many existing literatures involving incomplete hybrid information systems did not fully consider the impact of different conditional attributes on decision attributes and lacked effective aggregation methods to integrate weights and distances, we propose a new three-way decision method to deal with incomplete hybrid information systems with the help of triangular norms. First, in incomplete hybrid information systems, we redefine the distance between two objects based on conditional attributes and give the calculation formula of different data attributes in conditional attributes. At the same time, we define a new weight calculation method based on the relationship between conditional attributes and decision attributes. And then, by the distance between different conditional attributes and their corresponding weights, the hybrid distance is obtained using triangular norms. Furthermore, we use the hybrid distance to get the tolerance relation on the target set of an incomplete hybrid information system, thereby, using this tolerance relation, we get a new decision theoretic rough set model and the corresponding decision rules. Finally, by two experiments involving medical diagnosis, we demonstrate that our model has better classification ability, lower misclassification rate, and better stability compared to other corresponding models, thereby confirming that the proposed model provides a new and effective method for handling incomplete hybrid information systems in practical applications. • Novel method of handling incomplete hybrid data by three-way decision is proposed. • Novel distance formulas between different data types are proposed. • Method of calculating weight of condition attributes is obtained. • Method of combining distance and weight using triangular norms is obtained. [ABSTRACT FROM AUTHOR]

In recent years, there have been notable advancements in the theoretical research of existing meta-heuristic algorithms and in the development of novel algorithms. However, the research on the definition and construction methods of the so-called "mutation" operator, a concept widely used in existing meta-heuristic algorithms, remains notably insufficient. In addition, how to directly construct a meta-heuristic algorithm under the guidance of stability and invariance analysis continues to pose a challenge that has yet to resolve. This paper carefully investigates and rigorously defines the concepts of "mutation operator" and "mutation-driven meta-heuristic algorithm" as found in the literature. According to these definitions, a novel meta-heuristic algorithm, called Multi-direction guided Mutation-driven Stable swarm Intelligence algorithm with Translation and Rotation invariance (M2SITRI) is proposed, and the corresponding mutation operator is carefully designed based on the analysis of translation invariance, rotation invariance, order-1 stability, and order-2 stability. The stability and invariance properties of the M2SITRI algorithm are validated through numerical simulations, and its superior solving performance is demonstrated by benchmarking it against several other well-known existing meta-heuristic algorithms across a variety of test problems. • Mutation operator and mutation-driven meta-heuristic algorithm concepts are defined. • A novel translation and rotationally invariant meta-heuristic algorithm termed M2SITRI is proposed. • The invariance and stability properties and the excellent solving performance of M2SITRI algorithm are verified. [ABSTRACT FROM AUTHOR]

In recent years, Graph Neural Networks (GNNs) have become a powerful tool for graph representation in recommendation systems. Graphormer integrates its graph structure into a Transformer to provide state-of-the-art performance for graph prediction tasks. Meanwhile, self-supervised contrastive learning has succeeded in processing highly sparse data. Despite these successes, most graph contrastive methods perform random enhancement, such as node/edge perturbation, on a user-item interaction graph, or perform heuristic enhancement techniques, such as user clustering, to generate a new contrastive view. These methods cannot preserve the inherent semantic structure well and are susceptible to noise. In this paper, we propose a Graphormer-based graph contrastive learning method, GO-GCL, which reduces the influence of noise while retaining the inherent semantic structure to some extent, and improves the versatility and robustness of the recommendation based on contrastive learning. The model is represented by Graphormer and then uses singular value decomposition for contrastive enhancement. The experiments on Yelp datasets show that our model has a significant improvement over existing GCL-based approaches. In-depth analyses show the superiority and robustness of GO-GCL in data sparsity and prevalence bias. [Display omitted] • Propose a Graphormer-based contrastive learning paradigm, GO-GCL. • Alleviates the problems associated with inaccurate contrast signals. • Extensive experiments on Yelp demonstrate the performance advantages of GO-GCL. • Further analysis proves the rationality and robustness of the GO-GCL algorithm. [ABSTRACT FROM AUTHOR]

With the advancement of remote sensing technology, the amount of remote sensing image (RSI) has increased sharply. The explosion in data volume requires higher standards of image compression and encryption technology. This paper proposes a lossless and lossy encryption-compression method for RSI. Because RSIs contain important and unimportant information, using a normal algorithm could ignore their important information or reduce the computational efficiency, so we use the trained model by DeepLabv3+ to segment the region of interest (ROI) and the region of non-interest (RONI), and implement lossless and lossy algorithms respectively. The whole algorithm is based on the newly proposed hyperchaotic system 2D Tent coupled Infinite collapse map (2D TICM). According to the 2D TICM, the dynamic 3D Latin cube is generated. This cube is then used for dynamic scrambling algorithms between multiple planes (3D LMBS), within planes (3D LMIS), and for the multiple dynamic S boxes substitution algorithm. Due to the large size of RSI, the image is divided into blocks, and MATLAB parallel mechanism is used to encrypt each block at the same time. First, the lossy part is scrambled by 3D LMIS, then compressed with 2D compressive sensing (2D CS). Finally, each block is substituted with a different S box. A CS reconstruction algorithm combining decryption and reconstruction, 2D projection gradient chaotic decryption algorithm (2D PG-CD), is proposed. The lossless part embeds the encryption into JPEG-LS, encrypts while compressing, and then encrypts according to 3D LMBS and S boxes. The algorithm in this paper considers the characteristics of RSI, experiments show that it has good security and compression performance. • A hyperchaotic system 2D Tent coupled infinite collapse map (2D-TICM) is proposed. • Lossy compression 2D CS and 2D PG-CD improves the computational efficiency. • Encryption step is integrated into JPEG-LS and compression to retain important information. • Dynamic 3D Latin cube is designed with chaos and multiple S boxes substitution are proposed. [ABSTRACT FROM AUTHOR]

Transfer learning improves remaining useful life (RUL) prediction accuracy across domains by aligning data distributions for different operating conditions. However, the uncertainty caused by the complex working conditions and stochastic degradation process of rolling bearings is not considered, leading to poor credibility of the prediction results and affecting the development of the predictive maintenance strategy. In response to this problem, the paper proposes a deep subdomain adaptation time-quantile regression network (DSATQRN) model to compress rolling bearing uncertainty intervals of RUL across prediction. The model uses deep subdomain adaptation to align the feature distribution and introduces temporal correlation to construct a temporal quantile regression network to obtain interval prediction results. Finally, the uncertainty in the prediction results is quantified by kernel density estimation. The model was experimented with using the open XJTU-SY Bearing Datasets and IEEE PHM 2012 Challenge Datasets. It verifies the performance of the proposed model from three aspects: point prediction accuracy, interval prediction suitability, and probabilistic prediction overall performance. The experimental results show that the average interval coverage of the proposed model on the two datasets is 91.25% and 90.43%, and the average prediction interval width is 16.65% and 13.69%, respectively. It is demonstrated that the cross-domain prediction results of the DSATQRN possess high prediction accuracy and narrow prediction interval, and the model has good robustness and reliability. [Display omitted] • A cross-domain probability prediction model is built for rolling bearings. • The model compresses the uncertainty interval of RUL cross-domain prediction. • The model improves the reliability of the prediction results. • The model is end-to-end without manual feature extraction. [ABSTRACT FROM AUTHOR]

The 3D point cloud, a common format of 3D data, finds extensive applications in fields like remote sensing, surveying, robotics, and more. Addressing the challenges posed by insufficient weighted feature information and the offset between weighted results and task expectations in the attention mechanism, this paper proposes a Multi-scale Learnable Key-channel Attention Network (MLKNet). First, we introduce a feature feedback-repair module to mitigate the impact of information loss in the feature embedding process. This module aims to fully embed the original data into a high-dimensional feature space, ensuring a rich supply of feature information for subsequent transformer modules. Second, an efficient hierarchical local feature encoder extract and aggregate local features from point clouds at various scales, thereby significantly enhancing the model's capability to represent geometric structures. Third, a novel learnable key-channel attention module allows tasks to influence the feature selection and weighting process directly, make the highlighted features as close to task expectations as possible, effectively enhancing the network's perception of global semantic information. Our method was benchmarked on various tasks where we achieved overall accuracy (OA) of 92.3% on the ModelNet40 classification task and achieved instance mean intersection over union (ins. mIoU) of 87.6% on the ShapeNet-part segmentation task. The results indicate the superior performance of our method. • A feature feedback-repair module is used to reduce the loss during data embedding. • A hierarchical local feature encoder is used to aggregate the local features. • A novel matrix definition method is designed in learnable key-channel attention. [ABSTRACT FROM AUTHOR]

Striking a balance between diverse constraints and conflicting objectives is one of the most crucial issues in solving constrained multi-objective optimization problems (CMOPs). However, it remains challenging to existing methods, due to the reduced search space caused by the constraints. For this issue, this paper proposes a Pareto fronts relationship identification-based two-stage constrained evolutionary algorithm called RITEA, which balances objective optimization and constraint satisfaction by identifying and utilizing the relationship between the unconstrained Pareto front (UPF) and the constrained Pareto front (CPF). Specifically, the evolutionary process is divided into two collaborative stages: training stage and reinforcement stage. In the training stage, a relationship identification method is developed to estimate the relationship between UPF and CPF, which guides the population search direction. In the reinforcement stage, the corresponding evolutionary strategies are designed based on the identified relationship to enhance the accurate search on the CPF. Furthermore, a dynamic preference fitness function (termed DPF) is designed to adaptively maintain the balance of search preference between convergence and diversity. Compared to seven state-of-the-art algorithms on 36 benchmark CMOPs in three popular test suites, RITEA obtains 77.8% of the best IGD values and 66.7% of the best HV values. The experimental results show that RITEA exhibits highly competitively when dealing with CMOPs. [Display omitted] • RITEA, a two-stage constrained evolutionary algorithm for CMOPs. • Identify and utilize the relationship between UPF and CPF to balance optimization and constraint satisfaction. • Utilizes collaborative training stage and reinforcement stage to guide population search accurately. • Introduction of a dynamic preference fitness function to adaptively balance the search preferences between convergence and diversity. [ABSTRACT FROM AUTHOR]

Automated feature extraction from MRI brain scans and diagnosis of Alzheimer's disease are ongoing challenges. With advances in 3D imaging technology, 3D data acquisition is becoming more viable and efficient than its 2D counterpart. Rather than using feature-based vectors, in this paper, for the first time, we suggest a pipeline to extract novel covariance-based descriptors from the cortical surface using the Ricci energy optimization. The covariance descriptors are components of the nonlinear manifold of symmetric positive-definite matrices, thus we focus on using the Gaussian radial basis function to apply manifold-based classification to the 3D shape problem. Applying this novel signature to the analysis of abnormal cortical brain morphometry allows for diagnosing Alzheimer's disease. Experimental studies performed on about two hundred 3D MRI brain models, gathered from Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrate the effectiveness of our descriptors in achieving remarkable classification accuracy. [ABSTRACT FROM AUTHOR]