Search

Highlights • The robustness against malicious attacks and cascading failures are considered simultaneously. • An effective normalized robustness measure has been proposed. • Optimization algorithm based on multi-agent genetic algorithm has been devised and validated. • The obtained results may provide candidates to solve realistic dilemmas as reported in references. Abstract Networked systems exist broadly in daily life and have been greatly emphasized in recent studies. Since these networks are always exposed in complicated environments and may become damaged from different kinds of attacks and failures, including intentional attacks and cascading failures, many researchers have studied the models and mitigation strategies towards these two issues separately. As has been indicated by some reports and materials, the design of comprehensively robust networks against both intentional attacks and cascading failures is of significance in daily life but existing optimization methods seem to be helpless when dealing with this problem. Therefore, this paper focuses on designing such networks. A normalized measure is first designed to evaluate the comprehensive robustness of networks. Then, a multi-agent genetic algorithm is devised to search for robust structures initialized by an input network, termed MAGA-Net R. Combining local and global optima information, the proposed algorithm is validated to be effective on synthetic and real-world networks and to outperform existing optimization methods. The optimal network designed by the proposed algorithm could facilitate theoretical analyses on networked systems and may provide potential candidates to solve some realistic dilemmas. [ABSTRACT FROM AUTHOR]

A unified framework is proposed to select features by optimizing computationally feasible approximations of high-dimensional conditional mutual information (CMI) between features and their associated class label under different assumptions. Under this unified framework, state-of-the-art information theory based feature selection algorithms are re-derived, and a new algorithm is proposed to select features by optimizing a lower bound of the CMI with a weaker assumption than those adopted by existing methods. The new feature selection method integrates a plug-in component to distinguish redundant features from irrelevant ones for improving the feature selection robustness. Furthermore, a novel metric is proposed to evaluate feature selection methods based on simulated data. The proposed method has been compared with state-of-the-art feature selection methods based on the new evaluation metric and classification performance of classifiers built upon the selected features. The experiment results have demonstrated that the proposed method could achieve promising performance in a variety of feature selection problems. [ABSTRACT FROM AUTHOR]

In meta-heuristic optimisation, the robustness of a particular solution can be confirmed by re-sampling, which is reliable but computationally expensive, or by reusing neighbourhood solutions, which is cheap but unreliable. This work proposes a novel metric called the confidence measure to increase the reliability of the latter method, defines new confidence-based operators for robust meta-heuristics, and establishes a new robust optimisation approach called confidence-based robust optimisation. The confidence metric and five confidence-based operators are proposed and employed to design two new meta-heuristics: confidence-based robust Particle Swarm Optimisation and confidence-based robust Genetic Algorithm. A set of fifteen robust benchmark problems is employed to investigate the efficiencies of the proposed algorithms. The results show that the proposed metric is able to calculate the confidence level of solutions effectively during the optimisation process. In addition, the results demonstrate that the proposed operators can be employed to design a confident robust optimisation process and are readily applicable to different meta-heuristics. [ABSTRACT FROM AUTHOR]

Abstract: Sentence clustering has been successfully applied in document summarization to discover the topics conveyed in a collection of documents. However, existing clustering-based summarization approaches are seldom targeted for both diversity and coverage of summaries, which are believed to be the two key issues to determine the quality of summaries. The focus of this work is to explore a systematic approach that allows diversity and coverage to be tackled within an integrated clustering-based summarization framework. Given the fact that normally each topic can be described by a set of keywords and the choice of the keywords among the topics is topic-dependent, we take the advantage of the newly emerged subspace clustering to enable the flexibility of keyword selection and the improved quality of sentence clustering. On this basis, we develop two clustering-based optimization strategies, namely local optimization and global optimization to pursue our targets. Experimental results on the DUC datasets demonstrate effectiveness and robustness of the proposed approach. [Copyright &y& Elsevier]

Highlights: [•] We propose a temporally stable and noise robust matrix completion model. [•] A trust data detection method will boost the automation and performance of our method. [•] The optimization method for solving the proposed model is efficient and robust. [•] The parameters of our model are insensitive for a large range of values. [•] The outputs of our method are much more stable than the other methods. [Copyright &y& Elsevier]

Abstract: Artificial bee colony (ABC) algorithm is a relatively new optimization technique which has been shown to be competitive to other population-based algorithms. However, there is still an insufficiency in ABC regarding its solution search equation, which is good at exploration but poor at exploitation. To address this concerning issue, in this paper, we propose a novel ABC method called as EABC to improve the performance of ABC. In our method, in order to balance the exploration and the exploitation, two new search equations are presented to generate candidate solutions in the employed bee phase and the onlookers phase, respectively. Additionally, we use a more robust calculation to determine and compare the quality of alternative solutions. Experiments are conducted on a set of 48 benchmark functions and also two engineering optimization problems. The results show that EABC significantly improves the performance of ABC, offering faster global convergence, higher solution quality, and stronger robustness when compared with the other algorithms. [Copyright &y& Elsevier]

Abstract: The Harmony Search (HS) algorithm is a population-based metaheuristic optimisation algorithm. This algorithm is inspired by the music improvisation process in which the musician searches for harmony and continues to polish the pitches to obtain a better harmony. Although several variants of the HS algorithm have been proposed, their effectiveness in dealing with diverse problems is still unsatisfactory. The performances of these variants mainly depend on the selection of different parameters of the algorithm. In this paper, a new variant of the HS algorithm is proposed that maintains a proper balance between diversification and intensification throughout the search process by automatically selecting the proper pitch adjustment strategy based on its Harmony Memory. However, the performance of the proposed Intelligent Tuned Harmony Search (ITHS) algorithm is influenced by other parameters, such as the Harmony Memory Size (HMS) and the Harmony Memory Considering Rate (HMCR). The effects that varying these parameters have on the performance of the ITHS algorithm is also analysed in detail. The performance of the proposed ITHS algorithm is investigated and compared with eight state-of-the-art HS variants over 17 benchmark functions. Furthermore, to investigate the robustness of the proposed algorithm at higher dimensions, a scalability study is also performed. Finally, the numerical results obtained reflect the superiority of the proposed ITHS algorithm in terms of accuracy, convergence speed, and robustness when compared with other state-of-the-art HS variants. [Copyright &y& Elsevier]

Abstract: Several types of evolutionary computing methods are documented in the literature and are well known for solving unconstrained optimisation problems. This paper proposes a hybrid scheme that combines the merits of a global search algorithm, the shuffled frog-leaping algorithm (SFLA) and local exploration, extremal optimisation (EO) and that exhibits strong robustness and fast convergence for high-dimensional continuous function optimisation. A modified shuffled frog-leaping algorithm (MSFLA) is investigated that improves the leaping rule by properly extending the leaping step size and adding a leaping inertia component to account for social behaviour. To further improve the local search ability of MSFLA and speed up convergence, we occasionally introduce EO, which has an excellent local exploration capability, in the local exploration process of the MSFLA. It is characterised by alternating the coarse-grained Cauchy mutation and the fine-grained Gaussian mutation. Compared with standard particle swarm optimisation (PSO), SFLA and MSFLA for six widely used benchmark examples, the hybrid MSFLA-EO is shown to be a good and robust choice for solving high-dimensional continuous function optimisation problems. It possesses excellent performance in terms of the mean function values, the success rate and the fitness function evaluations (FFE), which is a rough measure of the complexity of the algorithm. [Copyright &y& Elsevier]

Abstract: In practical optimization problems, disturbances to a given instance are unavoidable due to unpredictable events which can occur when the system is running. In order to face these situations, many approaches have been proposed during the last years in the area of robust optimization. The basic idea of robustness is to provide a solution which can be used even if the input instance is disturbed, at the cost of optimality. However, the notion of robustness in every day life is much broader than that pursued in the area of robust optimization so far. In practice it is reasonable to consider a solution as robust, if a recovery strategy is available that can be applied when disturbing events occur in order to adapt the solution to the new situation. This suggests to study robustness and recoverability in a unified framework. Recently, a first tentative of unifying the notions of robustness and recoverability into a new integrated notion of recoverable robustness has been done in the context of railway optimization, see . Although this model represents a significant improvement, it has the following drawback: typically there is not only one disruption, but many of them might appear one after another. In this case, the solutions provided within the recoverable robustness are not satisfying. In this paper we extend the concept of recoverable robustness to deal not only with one single recovery step, but with many recovery steps. To this aim, we introduce the new notion of multi-stage recoverable robustness. We apply the new model in the context of timetabling and delay management problems. [Copyright &y& Elsevier]

Abstract: Practical production scheduling problems usually involve some “bottleneck” machines, the scheduling policies for which could significantly affect the final solution quality. Therefore, it is beneficial to identify the bottleneck machines beforehand, so that we can intensify the optimization for these machines in the subsequent solving stage. To this end, a bottleneck machine identification algorithm is proposed in this paper for the job shop scheduling problem with the objective of minimizing total tardiness. In order to obtain the instance-specific information about bottleneck machine distribution, we first propose a new optimization model which relaxes some conventional constraints of the standard job shop problem. Then, a simulated annealing algorithm is applied to solve this newly defined problem. Based on the optimization result, the bottleneck characteristic value (which is a measure of bottleneck level) is calculated for each machine. To utilize the obtained bottleneck information for job shop scheduling, we design a genetic algorithm which allocates more computational resources to the identified bottleneck machines by using a hybrid encoding scheme. Computational results verify the effectiveness and the robustness of the proposed bottleneck identification procedure. It is shown that intensifying the local search effort for the bottleneck machines will generally result in higher solution quality within reasonably short computational time. [Copyright &y& Elsevier]

Abstract: The performance of a kernel method often depends mainly on the appropriate choice of a kernel function. In this study, we present a data-dependent method for scaling the kernel function so as to optimize the classification performance of kernel methods. Instead of finding the support vectors in feature space, we first find the region around the separating boundary in input space, and subsequently scale the kernel function correspondingly. It is worth noting that the proposed method does not require a training step to enable a specified classification algorithm to find the boundary and can be applied to various classification methods. Experimental results using both artificial and real-world data are provided to demonstrate the robustness and validity of the proposed method. [Copyright &y& Elsevier]

Abstract: In accordance with the multi-objective nature of image watermarking, an optimal image watermarking approach using a multi-objective genetic algorithm is presented in this paper. Both watermarking parameters and embedding positions are often important factors affecting the performance of watermarking systems. The proposed multi-objective watermarking method can automatically optimize system parameters, and a variable-length mechanism is specially designed to search the most suitable positions for embedding watermarks. The method can also remove the difficult issue of determining optimal watermarking parameters from previous watermarking algorithms. The proposed multi-objective watermarking method directly deals with the problem of optimizing watermarking under non-dominated meaning, thus it can effectively avoid the difficulty of determining the optimally weighted factor in existing single-objective watermarking schemes. In addition, a Pareto-optimal set generated by multi-objective optimization can provide flexibility in selecting the most suitable watermarking parameters according to practical requirements. [Copyright &y& Elsevier]

Abstract: Swarm intelligence is a relatively novel field. It addresses the study of the collective behaviors of systems made by many components that coordinate using decentralized controls and self-organization. A large part of the research in swarm intelligence has focused on the reverse engineering and the adaptation of collective behaviors observed in natural systems with the aim of designing effective algorithms for distributed optimization. These algorithms, like their natural systems of inspiration, show the desirable properties of being adaptive, scalable, and robust. These are key properties in the context of network routing, and in particular of routing in wireless sensor networks. Therefore, in the last decade, a number of routing protocols for wireless sensor networks have been developed according to the principles of swarm intelligence, and, in particular, taking inspiration from the foraging behaviors of ant and bee colonies. In this paper, we provide an extensive survey of these protocols. We discuss the general principles of swarm intelligence and of its application to routing. We also introduce a novel taxonomy for routing protocols in wireless sensor networks and use it to classify the surveyed protocols. We conclude the paper with a critical analysis of the status of the field, pointing out a number of fundamental issues related to the (mis) use of scientific methodology and evaluation procedures, and we identify some future research directions. [Copyright &y& Elsevier]

Abstract: Differential Evolution (DE) is arguably one of the most powerful stochastic real-parameter optimization algorithms of current interest. DE operates through the similar computational steps as employed by a standard Evolutionary Algorithm (EA). However, unlike the traditional EAs, the DE-variants perturb the current-generation population members with the scaled differences of randomly selected and distinct population members. Therefore, no separate probability distribution has to be used, which makes the scheme self-organizing in this respect. Scale Factor (F) and Crossover Rate (Cr) are two very important control parameters of DE since the former regulates the step-size taken while mutating a population member in DE and the latter controls the number of search variables inherited by an offspring from its parent during recombination. This article describes a very simple yet very much effective adaptation technique for tuning both F and Cr, on the run, without any user intervention. The adaptation strategy is based on the objective function value of individuals in the DE population. Comparison with the best-known and expensive variants of DE over fourteen well-known numerical benchmarks and one real-life engineering problem reflects the superiority of proposed parameter tuning scheme in terms of accuracy, convergence speed, and robustness. [Copyright &y& Elsevier]

Abstract: Digital watermarking is a useful solution for digital rights management systems, and it has been a popular research topic in the last decade. Most watermarking related literature focuses on how to resist deliberate attacks by applying benchmarks to watermarked media that assess the effectiveness of the watermarking algorithm. Only a few papers have concentrated on the error-resilient transmission of watermarked media. In this paper, we propose an innovative algorithm for vector quantization (VQ) based image watermarking, which is suitable for error-resilient transmission over noisy channels. By incorporating watermarking with multiple description coding (MDC), the scheme we propose to embed multiple watermarks can effectively overcome channel impairments while retaining the capability for copyright and ownership protection. In addition, we employ an optimization technique, called tabu search, to optimize both the watermarked image quality and the robustness of the extracted watermarks. We have obtained promising simulation results that demonstrate the utility and practicality of our algorithm. [Copyright &y& Elsevier]

Abstract: This study presents a robust anti-windup fuzzy control approach for uncertain nonlinear time-delay systems with actuator saturations. The discussed system dynamics is presented by the Takagi–Sugeno (T–S) fuzzy model. To facilitate the design process, the nonlinearity of input saturation is characterized by a specific sector condition. Given a stabilizing dynamic output feedback fuzzy controller, an anti-windup control approach is then developed to maximize the size of estimate of the domain of attraction. Significantly, the convergence of all admissible initial states within this region could be ensured. Based on the Lyapunov–Krasovskii delay-independent and delay-dependent analyses, sufficient conditions for the robust stabilization are derived. These conditions are formulated as a convex optimization problem with constraints represented by a set of linear matrix inequalities, allowing the fuzzy controller to be synthesized more efficiently. Finally, numeric simulations are given to validate the proposed approach. [Copyright &y& Elsevier]

Abstract: Evolutionary computation plays a principal role in intelligent design automation. Evolutionary approaches have discovered novel and patentable designs. Nonetheless, evolutionary techniques may lead to designs that lack robustness. This critical issue is strongly connected to the concept of evolvability. In nature, highly evolvable species tend to be found in rapidly changing environments. Such species can be considered robust against environmental changes. Consequently, to create robust engineering designs it could be beneficial to use variable, rather than fixed, fitness criteria. In this paper, we study the performance of an evolutionary programming algorithm with periodical switching between goals, which are selected randomly from a set of related goals. It is shown by a dual-objective filter optimization example that altering goals may improve evolvability to a fixed goal by enhancing the dynamics of solution population, and guiding the search to areas where improved solutions are likely to be found. Our reference algorithm with a single goal is able to find solutions with competitive fitness, but these solutions are results of premature convergence, because they are poorly evolvable. By using the same algorithm with switching goals, we can extend the productive search length considerably; both the fitness and robustness of such designs are improved. [Copyright &y& Elsevier]