Bechikh, Slim

Elarbi M., Bechikh S., Coello Coello C.A.

Jerbi M., Chelly Dagdia Z., Bechikh S., Said L.B.

Malicious apps use a variety of methods to spread infections, take over computers and/or IoT devices, and steal sensitive data. Several detection techniques have been proposed to counter these attacks. Despite the promising results of recent malware detection strategies, particularly those addressing evolving threats, inefficiencies persist due to potential inconsistency in both the generated malicious malware and the pre-specified detection rules, as well as their crisp decision-making process. In this paper, we propose to address these issues by (i) considering the detection rules generation process as a Bi-Level Optimization Problem, where a competition between two levels (an upper level and a lower one) produces a set of effective detection rules capable of detecting new variants of existing and even unseen malware patterns. This bi-level strategy is subtly inspired by natural evolutionary processes, where organisms adapt and evolve through continuous interaction and competition within their environments. Furthermore, (ii) we leverage the fundamentals of Rough Set Theory, which reflects cognitive decision-making processes, to assess the true nature of artificially generated malicious patterns. This involves retaining only the consistent malicious patterns and detection rules and categorizing these rules into a three-way decision framework comprising accept, abstain, and reject options. Our novel malware detection technique outperforms several state-of-the-art methods on various Android malware datasets, accurately predicting new apps with a 96.76% accuracy rate. Moreover, our approach is versatile and effective in detecting patterns applicable to a variety of cybersecurity threats.

Ben Ali K., Louati H., Bechikh S.

The dynamic job shop scheduling problem (DJSSP) is an NP-hard optimization challenge, characterized by unpredictable events such as new job arrivals during scheduling. Our goal is to enhance search efficiency by integrating a learning system into the Particle Swarm Optimization (PSO) algorithm. This integration involves updating the inertia weight and incorporating a local search to refine search directions, termed PSO-IWLS (PSO with Inertia Weight Local Search). The PSO-IWLS has demonstrated superior performance compared to state-of-the-art approaches on large-scale benchmarks, excelling in both computation time and solution quality.

Mahouachi S., Elarbi M., Sethom K., Bechikh S., Coello C.A.

Ali K.B., Bechikh S., Louati A., Louati H., Kariri E.

Louati H., Louati A., Bechikh S., Kariri E.

Deep neural networks, specifically deep convolutional neural networks (DCNNs), have been highly successful in machine learning and computer vision, but a significant challenge when using these networks is choosing the right hyperparameters. As the number of layers in the network increases, the search space also becomes larger. To overcome this issue, researchers in deep learning have suggested using deep compression techniques to decrease memory usage and computational complexity. In this paper, we present a new approach for compressing deep CNNs by combining filter and channel pruning methods based on Evolutionary Algorithms (EA). This method involves eliminating filters and channels in order to decrease the number of parameters and computational complexity of the model. Additionally, we propose a bi-level optimization problem that interacts between the hyperparameters of the convolution layer. Bi-level optimization problems are known to be difficult as they involve two levels of optimization tasks, where only the optimal solutions to the lower-level problem are considered as feasible candidates for the upper-level problem. In this work, the upper-level problem is represented by a set of filters to be pruned in order to minimize the number of selected filters, while the lower-level problem is represented by a set of channels to be pruned in order to minimize the number of selected channels per filter. Our research has focused on developing a new method for solving bi-level problems, which we have named Bi-CNN-Pruning. To achieve this, we have adopted the Co-Evolutionary Migration-Based Algorithm (CEMBA) as our search engine. The Bi-CNN-Pruning method is then evaluated using image classification benchmarks on well-known datasets such as CIFAR-10 and CIFAR-100. The results of our evaluation demonstrate that our bi-level proposal outperforms state-of-the-art architectures, and we provide a detailed analysis of the results using commonly employed performance metrics.

Louati H., Louati A., Kariri E., Bechikh S.

Recent advancements in Computer Vision have opened up new opportunities for addressing complex healthcare challenges, particularly in the area of lung disease diagnosis. Chest X-rays, a commonly used radiological technique, hold great potential in this regard. To leverage this potential, researchers have proposed the use of deep learning methods for building computer-aided diagnostic systems. However, the design and compression of these systems remains a challenge, as it depends heavily on the expertise of the data scientists. To address this, we propose an automated method that utilizes an evolutionary algorithm (EA) to optimize the design and compression of a convolutional neural network (CNN) for X-Ray image classification. This method is capable of accurately classifying radiography images and detecting possible chest abnormalities and infections, including COVID-19. Additionally, the method incorporates transfer learning, where a pre-trained CNN model on a large dataset of chest X-ray images is fine-tuned for the specific task of detecting COVID-19. This approach can help to reduce the amount of labeled data required for the specific task and improve the overall performance of the model. Our method has been validated through a series of experiments against relevant state-of-the-art architectures.

Said R., Elarbi M., Bechikh S., Coello C.A., Said L.B.

Discretization-based feature selection approaches have shown interesting results when using several metaheuristic algorithms such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Ant Colony Optimization (ACO), etc. However, these methods share the same shortcoming which consists in encoding the problem solution as a sequence of cut-points. From this cut-points vector, the decision of deleting or selecting any feature is induced. Indeed, the number of generated cut-points varies from one feature to another. Thus, the higher the number of cut-points, the higher the probability of selecting the considered feature and vice versa. This fact leads to the deletion of possibly important features having a single or a low number of cut-points, such as the infection rate, the glycemia level, and the blood pressure. In order to solve the issue of the dependency relation between the feature selection (or removal) event and the number of its generated potential cut-points, we propose to model the discretization-based feature selection task as a bi-level optimization problem and then solve it using an improved version of an existing co-evolutionary algorithm, named I-CEMBA. The latter ensures the variation of the number of features during the migration process in order to deal with the multimodality aspect. The resulting algorithm, termed Bi-DFS (Bi-level Discretization-based Feature Selection), performs selection at the upper level while discretization is done at the lower level. The experimental results on several high-dimensional datasets show that Bi-DFS outperforms relevant state-of-the-art methods in terms of classification accuracy, generalization ability, and feature selection bias.

Salinas-Guerra R., Mezura-Montes E., Quiroz-Castellanos M., Mejía-de-Dios J., Bechikh S.

Said R., Bechikh S., Coello Coello C.A., Said L.B.

Jerbi M., Dagdia Z.C., Bechikh S., Said L.B.

Louati H., Louati A., Bechikh S., Kariri E.

Remarkable advancements have been achieved in machine learning and computer vision through the utilization of deep neural networks. Among the most advantageous of these networks is the convolutional neural network (CNN). It has been used in pattern recognition, medical diagnosis, and signal processing, among other things. Actually, for these networks, the challenge of choosing hyperparameters is of utmost importance. The reason behind this is that as the number of layers rises, the search space grows exponentially. In addition, all known classical and evolutionary pruning algorithms require a trained or built architecture as input. During the design phase, none of them consider the process of pruning. In order to assess the effectiveness and efficiency of any architecture created, pruning of channels must be carried out before transmitting the dataset and computing classification errors. For instance, following pruning, an architecture of medium quality in terms of classification may transform into an architecture that is both highly light and accurate, and vice versa. There exist countless potential scenarios that could occur, which prompted us to develop a bi-level optimization approach for the entire process. The upper level involves generating the architecture while the lower level optimizes channel pruning. Evolutionary algorithms (EAs) have proven effective in bi-level optimization, leading us to adopt the co-evolutionary migration-based algorithm as a search engine for our bi-level architectural optimization problem in this research. Our proposed method, CNN-D-P (bi-level CNN design and pruning), was tested on the widely used image classification benchmark datasets, CIFAR-10, CIFAR-100 and ImageNet. Our suggested technique is validated by means of a set of comparison tests with regard to relevant state-of-the-art architectures.

Chabbouh M., Bechikh S., Mezura-Montes E., Said L.B.

In multi-label classification, each instance could be assigned multiple labels at the same time. In such a situation, the relationships between labels and the class imbalance are two serious issues that should be addressed. Despite the important number of existing multi-label classification methods, the widespread class imbalance among labels has not been adequately addressed. Two main issues should be solved to come up with an effective classifier for imbalanced multi-label data. On the one hand, the imbalance could occur between labels and/or within a label. The “Between-labels imbalance” occurs where the imbalance is between labels however the “Within-label imbalance” occurs where the imbalance is in the label itself and it could occur across multiple labels. On the other hand, the labels’ processing order heavily influences the quality of a multi-label classifier. To deal with these challenges, we propose in this paper a bi-level evolutionary approach for the optimized induction of multivariate decision trees, where the upper-level role is to design the classifiers while the lower-level approximates the optimal labels’ ordering for each classifier. Our proposed method, named BIMLC-GA (Bi-level Imbalanced Multi-Label Classification Genetic Algorithm), is compared to several state-of-the-art methods across a variety of imbalanced multi-label data sets from several application fields and then applied on the miRNA-related diseases case study. The statistical analysis of the obtained results shows the merits of our proposal.

Boutaib S., Elarbi M., Bechikh S., Coello C.A., Said L.B.

Code smells (a.k.a. anti-patterns) are manifestations of poor design solutions that can deteriorate software maintainability and evolution. Existing works did not take into account the issue of uncertain class labels, which is an important inherent characteristic of the smells detection problem. More precisely, two human experts may have different degrees of uncertainty about the smelliness of a particular software class not only for the smell detection task but also for the smell type identification one. Unluckily, existing approaches usually reject and/or ignore uncertain data that correspond to software classes (i.e. dataset instances) with uncertain labels. Throwing away and/or disregarding the uncertainty factor could considerably degrade the detection/identification process effectiveness. From a solution approach viewpoint, there is no work in the literature that proposed a method that is able to detect and/or identify code smells while preserving the uncertainty aspect. The main goal of our research work is to handle the uncertainty factor, issued from human experts, in detecting and/or identifying code smells by proposing an evolutionary approach that is able to deal with anti-patterns classification with uncertain labels. We suggest Bi-ADIPOK, as an effective search-based tool that is capable to tackle the previously mentioned challenge for both detection and identification cases. The proposed method corresponds to an EA (Evolutionary Algorithm) that optimizes a set of detectors encoded as PK-NNs (Possibilistic K-nearest neighbors) based on a bi-level hierarchy, in which the upper level role consists on finding the optimal PK-NNs parameters, while the lower level one is to generate the PK-NNs. A newly fitness function has been proposed fitness function PomAURPC-OVA_dist (Possibilistic modified Area Under Recall Precision Curve One-Versus-All _distance, abbreviated PAURPC_d in this paper). Bi-ADIPOK is able to deal with label uncertainty using some concepts stemming from the Possibility Theory. Furthermore, the PomAURPC-OVA_dist is capable to process the uncertainty issue even with imbalanced data. We notice that Bi-ADIPOK is first built and then validated using a possibilistic base of smell examples that simulates and mimics the subjectivity of software engineers opinions. The statistical analysis of the obtained results on a set of comparative experiments with respect to four relevant state-of-the-art methods shows the merits of our proposal. The obtained detection results demonstrate that, for the uncertain environment, the PomAURPC-OVA_dist of Bi-ADIPOK ranges between 0.902 and 0.932 and its IAC lies between 0.9108 and 0.9407, while for the certain environment, the PomAURPC-OVA_dist lies between 0.928 and 0.955 and the IAC ranges between 0.9477 and 0.9622. Similarly, the identification results, for the uncertain environment, indicate that the PomAURPC-OVA_dist of Bi-ADIPOK varies between 0.8576 and 0.9273 and its IAC is between 0.8693 and 0.9318. For the certain environment, the PomAURPC-OVA_dist lies between 0.8613 and 0.9351 and the IAC values are between 0.8672 and 0.9476. With uncertain data, Bi-ADIPOK can find 35% more code smells than the second best approach (i.e., BLOP). Furthermore, Bi-ADIPOK has succeeded to reduce the number of false alarms (i.e., misclassified smelly instances) by 12%. In addition, our proposed approach can identify 43% more smell types than BLOP and reduces the number of false alarms by 32%. The same results have been obtained for the certain environment, demonstrating Bi-ADIPOK’s ability to deal with such environment. • Code smells detection could be an uncertain data classification problem. • The existing works have ignored the uncertainty existing in the class labels. • Ignoring the uncertain data could lead to the loss of information. • A novel Bilevel approach called Bi-ADIPOK is suggested to detect and/or identify code smells. • An experimental study is performed using the PAURPC_d and the IAC metrics.

Ben Amor O., Chelly Dagdia Z., Bechikh S., Ben Said L.

Recently, the efficient deployment of wireless sensor networks (WSNs) has become a leading field of research in WSN design optimization. Practical scenarios related to WSN deployment are often considered as optimization models with multiple conflicting objectives that are simultaneously enhanced. In the related literature, it had been shown that moving from mono-objective to multi-objective resolution of WSN deployment is beneficial. However, since the deployment of real-world WSNs encompasses more than three objectives, a multi-objective optimization may harm other deployment criteria that are conflicting with the already considered ones. Thus, our aim is to go further, explore the modeling and the resolution of WSN deployment in a many-objective (i.e., optimization with more than three objectives) fashion and especially, exhibit its added value. In this context, we first propose a many-objective deployment model involving seven conflicting objectives, and then we solve it using an adaptation of the Decomposition-based Evolutionary Algorithm “ $$\theta$$ -DEA”. The developed adaptation is named “WSN- $$\theta$$ -DEA” and is validated through a detailed experimental study.

Wang Z., Zhao H., Lu Q., Wu T.

Chen P., Liang J., Qiao K., Ban X., Suganthan P.N., Lin H., Zhang J.

AlOmar E.A.

Chen L., Cai M., Li Q.

Du Q., Guo W., Mo S., Wang Y., Qin T., Zhu B., Zheng H.

Sharma Y.K., Tomar D.S., Pateriya R.K., Bhandari S.

Yang D., Fan Q.

Gu Q., Chen Y., Wang D., Liu D.

Zhao R., ALDharhani G.S., Ratnavelu K., Thambiayya S., Jasser M.B., Majeed A.P., Luo Y.

Zhou Y., Chai Z., Li Y., Li J.

Guo J., Zhao Y., Zheng T., Chen Z., Jiang M., Ding Z.

Hao L., Peng W., Liu J., Zhang W., Li Y., Qin K.

Kovács S., Budai C., Botzheim J.

Abstract In this paper, we present the Colonial Bacterial Memetic Algorithm (CBMA), an advanced evolutionary optimization approach for robotic applications. CBMA extends the Bacterial Memetic Algorithm by integrating Cultural Algorithms and co-evolutionary dynamics inspired by bacterial group behavior. This combination of natural and artificial evolutionary elements results in a robust algorithm capable of handling complex challenges in robotics, such as constraints, multiple objectives, large search spaces, and complex models, while delivering fast and accurate solutions. CBMA incorporates features like multi-level clustering, dynamic gene selection, hierarchical population clustering, and adaptive co-evolutionary mechanisms, enabling efficient management of task-specific parameters and optimizing solution quality while minimizing resource consumption. The algorithm’s effectiveness is demonstrated through a real-world robotic application, achieving a 100% success rate in a robot arm’s ball-throwing task usually with significantly fewer iterations and evaluations compared to other methods. CBMA was also evaluated using the CEC-2017 benchmark suite, where it consistently outperformed state-of-the-art optimization algorithms, achieving superior outcomes in 71% of high-dimensional cases and demonstrating up to an 80% reduction in required evaluations. These results highlight CBMA’s efficiency, adaptability, and suitability for specialized tasks. Overall, CBMA exhibits exceptional performance in both real-world and benchmark evaluations, effectively balancing exploration and exploitation, and representing a significant advancement in adaptive evolutionary optimization for robotics.

Ghosh A., Srinivasan K.

Liu W., Zhu D., Zhou C., Cheng S.

Li K.

Xu Y., Li F., Zhang H., Li W.

The performance of evolutionary algorithms using reference vectors to guide the evolution process mainly depends on the adaptive reference vector update strategy. In order to solve the challenging many-objective optimization problems with irregular Pareto fronts, this paper proposes an adaptive reference vector update strategy based on the Pareto front density estimation, which estimates the true Pareto front by finding sparse regions while ensuring the uniform distribution of reference vectors. In addition, an improved environmental selection strategy using the angle-based neighborhood density estimation has been proposed for estimating the neighborhood density to effectively guide the population evolution. On this basis, this paper proposes an adaptive reference vector guided many-objective optimization algorithm based on Pareto front density estimation (MaOEA-PDE). Experimental results on a large number of benchmark problems show MaOEA-PDE achieves better performance compared with some state-of-the-art algorithms in the literature.

Wu X., Yan X., Guan D., Wei M.

The dynamic job-shop scheduling problem (DJSP) is a type of scheduling tasks where rescheduling is performed when encountering the uncertainties such as the uncertain operation processing time. However, the current deep reinforcement learning (DRL) scheduling approaches are hard to train convergent scheduling policies as the problem scale increases, which is very important for rescheduling under uncertainty. In this paper, we propose a DRL scheduling method for DJSP based on the proximal policy optimization (PPO) with hybrid prioritized experience replay. The job shop scheduling problem is formulated as a sequential decision-making problem based on Markov Decision Process (MDP) where a novel state representation is designed based on the feasible solution matrix which depicts the scheduling order of a scheduling task, a set of paired priority dispatching rules (PDR) are used as the actions and a new intuitive reward function is established based on the machine idle time. Moreover, a new hybrid prioritized experience replay method for PPO is proposed to reduce the training time where samples with positive temporal-difference (TD) error are replayed. Static experiments on classic benchmark instances show that the make-span obtained by our scheduling agent has been reduced by 1.59% on average than the best known DRL methods. In addition, dynamic experiments demonstrate that the training time of the reused scheduling policy is reduced by 27% compared with the retrained policy when encountering uncertainties such as uncertain operation processing time.

Zhu N., Gong G., Lu D., Huang D., Peng N., Qi H.

Order cancellation, due to such as customer plan adjustments or market changes, usually occurs in the real production environment of distributed flexible job shop scheduling problem (DFJSP). However, thus far, all exiting researches about DFJSP have not consider order cancellation, which normally leads to resource waste and makes the original scheme infeasible. Hence, in this work, we propose a DFJSP considering order cancellation (DFJSPC) for the first time; and design a reformative memetic algorithm (RMA) to solve the DFJSPC aiming at optimizing the makespan and total energy consumption. In the RMA, a five-layer encoding operator and a new load balancing initialization method are designed to improve the quality of the initial population. Some effective crossover, mutation and local search operators are designed, which can fully expand the solution space of the algorithm and improve its convergence speed. A total of 60 DFJSPC benchmark instances are constructed, and some comparative experiments are carried out among the proposed RMA and three well-known algorithms, namely NNIA, NSGA-II and NSGA-III. The final experimental results verified the outstanding performance of the RMA. This research will provide a theoretical basis for the order cancellation problem in distributed production settings, and help manufacturers to properly handle canceled orders to reduce resource waste and reschedule the infeasible schemes causing from order cancellation.

Shao X., Kshitij F.S., Kim C.S.

AbstractThe job shop scheduling problem (JSSP) is critical for building one smart factory regarding resource management, effective production, and intelligent supply. However, it is still very challenging due to the complex production environment. Besides, most current research only focuses on classical JSSP, while flexible JSSP (FJSSP) is more usual. This article proposes an effective method, GAILS, to deal with JSSP and FJSSP based on genetic algorithm (GA) and iterative local search (ILS). GA is used to find the approximate global solution for the JSSP instance. Each instance was encoded into machine and subtask sequences. The corresponding machine and subtasks chromosome could be obtained through serval-time gene selection, crossover, and mutation. Moreover, multi-objects, including makespan, average utilization ratio, and maximum loading, are used to choose the best chromosome to guide ILS to explore the best local path. Therefore, the proposed method has an excellent search capacity and could balance globality and diversity. To verify the proposed method's effectiveness, the authors compared it with some state-of-the-art methods on sixty-six public JSSP and FJSSP instances. The comparative analysis confirmed the proposed method's effectiveness for classical JSSP and FJSSP in makespan, average utilization ratio, and maximum loading. Primarily, it obtains optimal-like solutions for several instances and outperforms others in most instances.

Werth B., Karder J., Heckmann M., Wagner S., Affenzeller M.

Real-world production scheduling scenarios are often not discrete, separable, iterative tasks but rather dynamic processes where both external (e.g., new orders, delivery shortages) and internal (e.g., machine breakdown, timing uncertainties, human interaction) influencing factors gradually or abruptly impact the production system. Solutions to these problems are often very specific to the application case or rely on simple problem formulations with known and stable parameters. This work presents a dynamic scheduling scenario for a production setup where little information about the system is known a priori. Instead of fully specifying all relevant problem data, the timing and batching behavior of machines are learned by a machine learning ensemble during operation. We demonstrate how a meta-heuristic optimization algorithm can utilize these models to tackle this dynamic optimization problem, compare the dynamic performance of a set of established construction heuristics and meta-heuristics and showcase how models and optimizers interact. The results obtained through an empirical study indicate that the interaction between optimization algorithm and machine learning models, as well as the real-time performance of the overall optimization system, can impact the performance of the production system. Especially in high-load situations, the dynamic algorithms that utilize solutions from previous problem epochs outperform the restarting construction heuristics by up to ~24%.

Song L., Li Y., Xu J.

The dynamic job-shop scheduling problem is a complex and uncertain task that involves optimizing production planning and resource allocation in a dynamic production environment. Traditional methods are limited in effectively handling dynamic events and quickly generating scheduling solutions; in order to solve this problem, this paper proposes a solution by transforming the dynamic job-shop scheduling problem into a Markov decision process and leveraging deep reinforcement learning techniques. The proposed framework introduces several innovative components, which make full use of human domain knowledge and machine computing power, to realize the goal of man–machine collaborative decision-making. Firstly, we utilize disjunctive graphs as the state representation, capturing the complex relationships between various elements of the scheduling problem. Secondly, we select a set of dispatching rules through data envelopment analysis to form the action space, allowing for flexible and efficient scheduling decisions. Thirdly, the transformer model is employed as the feature extraction module, enabling effective capturing of state relationships and improving the representation power. Moreover, the framework incorporates the dueling double deep Q-network with prioritized experience replay, mapping each state to the most appropriate dispatching rule. Additionally, a dynamic target strategy with an elite mechanism is proposed. Through extensive experiments conducted on multiple examples, our proposed framework consistently outperformed traditional dispatching rules, genetic algorithms, and other reinforcement learning methods, achieving improvements of 15.98%, 17.98%, and 13.84%, respectively. These results validate the effectiveness and superiority of our approach in addressing dynamic job-shop scheduling problems.

Liu C., Huang T.

lv Z., Liao Z., Liu Y., Zhao J.

The by-product gas is an important secondary energy in the iron and steel industry. It is important to make the by-product gas’s utilization efficient and reasonable,which is the key to improve the economic efficiency and the level of energy conservation and emission reduction. Aiming at the problems of complex dynamic changes, difficult to accurately model and difficult to predict real-time traffic in gas energy system, this paper proposes an optimal scheduling method based on meta-learning multi-objective particle swarm optimization algorithm. The gas energy optimization scheduling problem is modeled as a multi-objective dynamic scheduling optimization problem. The difficulty of this problem is to comprehensively consider multiple indicators, continuously adapt to the change of the objective function over time, and track the changing Pareto optimal solution set. One of the promising solutions is to build prediction models with better performance under dynamic changes. The model we propose makes full use of the optimization results and the experience information of the optimization process in the historical optimization task. We propose a dynamic parameter initialization method based on meta-learning by starting multiple historical optimization tasks at the same time. Thus,the comprehensive parameters of different tasks are obtained, and a general and best base parameter is learned. Experiments show that the proposed model has better convergence and accuracy than the conventional algorithm.

Wang B., Singh H.K., Ray T.

Salvakkam D.B., Saravanan V., Jain P.K., Pamula R.

The increasing popularity of cloud computing systems has drawn significant attention from academics and businesses for several decades. However, cloud computing systems are plagued with several concerns, such as privacy, confidentiality, and availability, which can be detrimental to their performance. Intrusion detection has emerged as a critical issue, particularly in detecting new types of intrusions that can compromise the security of cloud systems. Preventive risk models have been developed to check the cloud for potential threats, and the rise of quantum computing attacks necessitates the deployment of an intrusion detection system (IDS) for cloud security risk assessment. This research proposes a unique method for detecting cloud computing intrusions by utilizing the KDDcup 1999, UNSW-NB15, and NSL-KDD datasets to address these concerns. This proposed system is designed to achieve two objectives. Firstly, it analyzes the disadvantages of existing IDS, and secondly, it presents an accuracy enhancement model of IDS. The proposed Ensemble Intrusion Detection Model for Cloud Computing Using Deep Learning (EICDL) is designed to detect intrusions effectively. The performance of the proposed model is compared to modern machine learning methods and existing IDS, and the experimental findings indicate that the EICDL ensemble technique improves detection and can identify subsequent attacks/intrusions with a recall rate of 92.14%. The proposed method EICDL ensemble technique significantly improves the accuracy and efficiency of intrusion detection in cloud systems.

Zhang Z., Gao S., Lei Z., Xiong R., Cheng J.

Fontes D.B., Homayouni S.M., Gonçalves J.F.

This work addresses a variant of the job shop scheduling problem in which jobs need to be transported to the machines processing their operations by a limited number of vehicles. Given that vehicles must deliver the jobs to the machines for processing and that machines need to finish processing the jobs before they can be transported, machine scheduling and vehicle scheduling are intertwined. A coordinated approach that solves these interrelated problems simultaneously improves the overall performance of the manufacturing system. In the current competitive business environment, and integrated approach is imperative as it boosts cost savings and on-time deliveries. Hence, the job shop scheduling problem with transport resources (JSPT) requires scheduling production operations and transport tasks simultaneously. The JSPT is studied considering the minimization of two alternative performance metrics, namely: makespan and exit time. Optimal solutions are found by a mixed integer linear programming (MILP) model. However, since integrated production and transportation scheduling is very complex, the MILP model can only handle small-sized problem instances. To find good quality solutions in reasonable computation times, we propose a hybrid particle swarm optimization and simulated annealing algorithm (PSOSA). Furthermore, we derive a fast lower bounding procedure that can be used to evaluate the performance of the heuristic solutions for larger instances. Extensive computational experiments are conducted on 73 benchmark instances, for each of the two performance metrics, to assess the efficacy and efficiency of the proposed PSOSA algorithm. These experiments show that the PSOSA outperforms state-of-the-art solution approaches and is very robust.

Louati H., Louati A., Bechikh S., Kariri E.

Remarkable advancements have been achieved in machine learning and computer vision through the utilization of deep neural networks. Among the most advantageous of these networks is the convolutional neural network (CNN). It has been used in pattern recognition, medical diagnosis, and signal processing, among other things. Actually, for these networks, the challenge of choosing hyperparameters is of utmost importance. The reason behind this is that as the number of layers rises, the search space grows exponentially. In addition, all known classical and evolutionary pruning algorithms require a trained or built architecture as input. During the design phase, none of them consider the process of pruning. In order to assess the effectiveness and efficiency of any architecture created, pruning of channels must be carried out before transmitting the dataset and computing classification errors. For instance, following pruning, an architecture of medium quality in terms of classification may transform into an architecture that is both highly light and accurate, and vice versa. There exist countless potential scenarios that could occur, which prompted us to develop a bi-level optimization approach for the entire process. The upper level involves generating the architecture while the lower level optimizes channel pruning. Evolutionary algorithms (EAs) have proven effective in bi-level optimization, leading us to adopt the co-evolutionary migration-based algorithm as a search engine for our bi-level architectural optimization problem in this research. Our proposed method, CNN-D-P (bi-level CNN design and pruning), was tested on the widely used image classification benchmark datasets, CIFAR-10, CIFAR-100 and ImageNet. Our suggested technique is validated by means of a set of comparison tests with regard to relevant state-of-the-art architectures.

Kariri E., Louati H., Louati A., Masmoudi F.

Artificial Neural Networks (ANNs) are machine learning algorithms inspired by the structure and function of the human brain. Their popularity has increased in recent years due to their ability to learn and improve through experience, making them suitable for a wide range of applications. ANNs are often used as part of deep learning, which enables them to learn, transfer knowledge, make predictions, and take action. This paper aims to provide a comprehensive understanding of ANNs and explore potential directions for future research. To achieve this, the paper analyzes 10,661 articles and 35,973 keywords from various journals using a text-mining approach. The results of the analysis show that there is a high level of interest in topics related to machine learning, deep learning, and ANNs and that research in this field is increasingly focusing on areas such as optimization techniques, feature extraction and selection, and clustering. The study presented in this paper is motivated by the need for a framework to guide the continued study and development of ANNs. By providing insights into the current state of research on ANNs, this paper aims to promote a deeper understanding of ANNs and to facilitate the development of new techniques and applications for ANNs in the future.