Differential Evolution (DE) algorithm is a well-known metaheuristic algorithm that features a simple structure and excellent optimization performance. However, it still suffers from premature convergence or stagnation...
详细信息
Differential Evolution (DE) algorithm is a well-known metaheuristic algorithm that features a simple structure and excellent optimization performance. However, it still suffers from premature convergence or stagnation when dealing with complex optimization problems. To avoid these dilemmas in the DE algorithm, we propose a novel DE variant, abbreviated as PISCDE, which is based on periodic intervention and strategic collaboration mechanisms. PISCDE incorporates two types of operations: routine operation and intervention operation. The routine operation employs two mutation strategies with different functional positions to drive the population toward the optimal position. In contrast, the intervention operation uses two intervention strategies with distinct functional roles to restore population diversity and is executed only when a fixed number of iterations is reached. Additionally, to achieve a better balance between global exploration and local exploitation during the optimization process, we propose several strategic collaboration mechanisms. These mechanisms are based on the positioning analysis of different strategies and the interaction analysis between strategies and their corresponding control parameters. To verify the optimization performance of PISCDE, we selected nine comparison algorithms with outstanding optimization performance that have been proposed in the last five years. We used the IEEE CEC 2014 testbed to construct comparative experiments. Based on the comparative results, three conclusions can be drawn: (1) PISCDE has the best overall optimization performance among all the algorithms. (2) PISCDE performs more significantly on complex test problems. (3) PISCDE shows more impressive optimization performance when the dimension of the test problems is increased.
Water distribution networks (WDNs) are essential for modern cities, as effective design can reduce construction costs and ensure reliable service. As cities expand, optimizing these large networks becomes increasingly...
详细信息
Water distribution networks (WDNs) are essential for modern cities, as effective design can reduce construction costs and ensure reliable service. As cities expand, optimizing these large networks becomes increasingly complex. In this work, we introduce a novel approach by combining simulated annealing (SA) with variable neighbourhood search (VNS) into a single heuristic algorithm for WDN optimization, marking the first use of the SA-VNS method in this context. Additionally, we apply Taguchi's design of experiments (DOE) to tune the parameters of the SA-VNS algorithm specifically for water networks. We tested our new algorithm on four standard benchmark networks and a real-world WDN, including a case study of a large city. Our results demonstrate that the SA-VNS algorithm outperforms existing methods in terms of cost, speed, and overall effectiveness, making this research a significant advancement in both heuristic methods and parameter-tuning techniques for WDN optimization.
In real-world optimisation, it is common to face several sub-problems interacting and forming the main problem. There is an inter-dependency between the sub-problems, making it impossible to solve such a problem by fo...
详细信息
Over the past few years, quantum-inspired genetic algorithm, as the forerunner of quantum evolutionary algorithms has made significant achievements in solving optimization problems in various fields. The ability of th...
详细信息
Over the past few years, quantum-inspired genetic algorithm, as the forerunner of quantum evolutionary algorithms has made significant achievements in solving optimization problems in various fields. The ability of this algorithm to comprehensively search the problem space and discover global extrema is due to the power of the exploration process in this algorithm, which in turn is attributed to the mutations resulting from quantum rotation gates. But the best results in evolutionary algorithms are obtained when there is a reasonable balance between the exploration process (general search) and the exploitation process (local search). Under such conditions, the convergence speed of the algorithm in finding the global optima is no longer suppressed. In this research, a memetic quantum-inspired evolutionary algorithm is proposed, which is a combination of a quantum genetic algorithm and a local search-based meta-heuristic called tabu search. In the proposed method, mutations of the rotation gate cover the entire problem space and access distant solutions. Consequently, the impulses generated by the tabu search in each iteration limit the range of motion and access to nearby solutions. Consecutive execution of these two processes turns the classical mutation operator into a directional mutation and accelerates the convergence of evolution. The evaluation results of this algorithm on unimodal and multimodal benchmark functions reflect the superiority of the proposed method over state-of-the-art methods in terms of convergence rate and running time.
Spiking Neural Networks (SNNs) constitute a representative example of neuromorphic computing in which event-driven computation is mapped to neuron spikes reducing power consumption. A challenge that limits the general...
详细信息
ISBN:
(纸本)9783031429200;9783031429217
Spiking Neural Networks (SNNs) constitute a representative example of neuromorphic computing in which event-driven computation is mapped to neuron spikes reducing power consumption. A challenge that limits the general adoption of SNNs is the need for mature training algorithms compared with other artificial neural networks, such as multilayer perceptrons or convolutional neural networks. This paper explores the use of evolutionary algorithms as a black-box solution for training SNNs. The selected SNN model relies on the Izhikevich neuron model implemented in hardware. Differently from state-of-the-art, the approach followed in this paper integrates within the same System-on-a-chip (SoC) both the training algorithm and the SNN fabric, enabling continuous network adaptation in-field and, thus, eliminating the barrier between offline (training) and online (inference). A novel encoding approach for the inputs based on receptive fields is also provided to improve network accuracy. Experimental results demonstrate that these techniques perform similarly to other algorithms in the literature without dynamic adaptability for classification and control problems.
JavaScript (JS) is nowadays the only language that can be used to develop web-based client-server applications in both tiers, client and server. This makes it an interesting choice for developing distributed evolution...
详细信息
ISBN:
(纸本)9781450328814
JavaScript (JS) is nowadays the only language that can be used to develop web-based client-server applications in both tiers, client and server. This makes it an interesting choice for developing distributed evolutionary computation experiments, but the best way from algorithmic and practical point of views is not clear, so we will compare different distributed EC architectures in JavaScript using NodEO, an open source JS framework released by us.
evolutionary computation tools are able to process real valued numerical sets in order to extract suboptimal solution of designed problem. Data clustering algorithms have been intensively used for image segmentation i...
详细信息
ISBN:
(纸本)9781629935201
evolutionary computation tools are able to process real valued numerical sets in order to extract suboptimal solution of designed problem. Data clustering algorithms have been intensively used for image segmentation in remote sensing applications. Despite of wide usage of evolutionary algorithms on data clustering, their clustering performances have been scarcely studied by using clustering validation indexes. In this paper, the recently proposed evolutionary algorithms (i.e., Artificial Bee Colony Algorithm (ABC), Gravitational Search Algorithm (GSA), Cuckoo Search Algorithm (CS), Adaptive Differential Evolution Algorithm (JADE), Differential Search Algorithm (DSA) and Backtracking Search Optimization Algorithm (BSA)) and some classical image clustering techniques (i.e., k-means, fcm, som networks) have been used to cluster images and their performances have been compared by using four clustering validation indexes. Experimental test results exposed that evolutionary algorithms give more reliable cluster-centers than classical clustering techniques, but their convergence time is quite long.
Multi-objective evolutionary algorithms (MOEAs) have been the subject of a large research effort over the past two decades. Traditionally, these MOEAs have been seen as monolithic units, and their study was focused on...
详细信息
ISBN:
(纸本)9783319107622;9783319107615
Multi-objective evolutionary algorithms (MOEAs) have been the subject of a large research effort over the past two decades. Traditionally, these MOEAs have been seen as monolithic units, and their study was focused on comparing them as blackboxes. More recently, a component wise view of MOEAs has emerged, with flexible frameworks combining algorithmic components from different MOEAs. The number of available algorithmic components is large, though, and an algorithm designer working on a specific application cannot analyze all possible combinations. In this paper, we investigate the automatic design of MOEAs, extending previous work on other multi-objective metaheuristics. We conduct our tests on four variants of the permutation flowshop problem that differ on the number and nature of the objectives they consider. Moreover, given the different characteristics of the variants, we also investigate the performance of an automatic MOEA designed for the multi-objective PFSP in general. Our results show that the automatically designed MOEAs are able to outperform six traditional MOEAs, confirming the importance and efficiency of this design methodology.
With the rapid development of next-generation mobile network services, there is a growing need for customized services to meet the demands of various network functions. Leveraging the Software-Defined Networking (SDN)...
详细信息
With the rapid development of next-generation mobile network services, there is a growing need for customized services to meet the demands of various network functions. Leveraging the Software-Defined Networking (SDN) architecture, Network Function Virtualization (NFV) enhances service delivery flexibility by virtualizing network appliances. This allows for Service Function Chain (SFC), which further enhances service delivery flexibility through centralized, programmable management. However, existing works require manual adjustments and tuning when adapting to evolving user demands and network expansions, lacking the flexibility needed for changing network conditions. With the rise of Large Language Models (LLMs), the automation of network management has gained new momentum by understanding programming logic, generating code, and incorporating advanced knowledge of network and optimization. This paper introduces an LLM-assisted network operating system framework and presents a case for LLM-assisted SFC optimization. Finally, it proposes an NSGA2-based multi-objective LLM optimization algorithm, which continuously updates the heuristic code policies through evolutionary iterations. Simulation results validate the effectiveness of this approach in achieving stable and efficient multi-objective optimization for SFC deployment.
暂无评论