Swarm and Evolutionary Computation最新文献_第2页

Solving dynamic multi-objective engineering design problems via fuzzy c-means prediction algorithm 用模糊c均值预测算法求解动态多目标工程设计问题

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-20 DOI: 10.1016/j.swevo.2025.102057

Qingyang Zhang , Xueliang Fu , Shengxiang Yang , Shouyong Jiang , Miqing Li , Zedong Zheng

{"title":"Solving dynamic multi-objective engineering design problems via fuzzy c-means prediction algorithm","authors":"Qingyang Zhang , Xueliang Fu , Shengxiang Yang , Shouyong Jiang , Miqing Li , Zedong Zheng","doi":"10.1016/j.swevo.2025.102057","DOIUrl":"10.1016/j.swevo.2025.102057","url":null,"abstract":"<div><div>This paper proposes a new prediction algorithm by integrating the fuzzy c-means and regression analysis fitting techniques with multi-objective differential evolution (FRMODE) to solve dynamic multi-objective optimization problems. When environmental changes are detected, the main purpose of FRMODE is to predict high-quality populations that can effectively track the moving Pareto-optimal set. Specifically, the fuzzy c-means (FCM) algorithm clusters the populations obtained from the past two adjacent environments. The center points of populations are utilized to define the moving direction, which is used to predict high-quality agents based on previous non-dominated individuals. Then, linear and non-linear regression analysis fitting strategies are developed to model the distribution of variables according to the variables’ characteristics. Besides that, the partial mutation strategy is also utilized to guide individuals toward more promising regions by intensifying the search around current agents. To evaluate the performance of the proposed algorithm, experiments are conducted on a set of benchmark functions with various dynamic difficulties, as well as on two classical dynamic engineering design problems. The experimental results demonstrate that FRMODE is more competitive compared with several state-of-the-art algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102057"},"PeriodicalIF":8.2,"publicationDate":"2025-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144665707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A dynamic task-assisted constrained multimodal multi-objective optimization algorithm based on reinforcement learning 基于强化学习的动态任务辅助约束多模态多目标优化算法

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-19 DOI: 10.1016/j.swevo.2025.102087

Zheng Wang , Qianlin Ye , Wanliang Wang , Guoqing Li , Rui Dai

{"title":"A dynamic task-assisted constrained multimodal multi-objective optimization algorithm based on reinforcement learning","authors":"Zheng Wang , Qianlin Ye , Wanliang Wang , Guoqing Li , Rui Dai","doi":"10.1016/j.swevo.2025.102087","DOIUrl":"10.1016/j.swevo.2025.102087","url":null,"abstract":"<div><div>Constrained multimodal optimization problems (CMMOPs) are required to satisfy constraint limitations and ensure the convergence and diversity of the solutions in the objective and decision spaces. It increases the difficulty of solving the optimization problems. To design efficient constrained multimodal multi-objective optimization evolutionary algorithms (CMMOEAs) to solve them is a hot topic today. A novel dynamic auxiliary task selection algorithm (DTCMMO-RL) is designed based on the multi-task framework and reinforcement learning. The algorithm designs three auxiliary tasks to optimize constrained multi-objective problems, simple multi-objective problems and multimodal optimization problems, respectively. At the same time, Q-learning in reinforcement learning is employed to dynamically select the current optimal auxiliary task to utilize the useful information obtained rationally. In addition, an indicator (IGDXp) capable of evaluating the comprehensive performance of the solutions in the objective space and decision space is designed. To verify the excellence of DTCMMO-RL, a series of experiments with 11 comparison algorithms on CMMF and CMMOP are conducted to verify the feasibility and effectiveness of multiple strategies.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102087"},"PeriodicalIF":8.2,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An offline data-driven process for learning operator selection from metaheuristic search traces 从元启发式搜索痕迹中学习算子选择的离线数据驱动过程

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-19 DOI: 10.1016/j.swevo.2025.102058

Panagiotis Kalatzantonakis, Angelo Sifaleras, Nikolaos Samaras

{"title":"An offline data-driven process for learning operator selection from metaheuristic search traces","authors":"Panagiotis Kalatzantonakis, Angelo Sifaleras, Nikolaos Samaras","doi":"10.1016/j.swevo.2025.102058","DOIUrl":"10.1016/j.swevo.2025.102058","url":null,"abstract":"<div><div>Trained Reward-based Action Classification Engine (TRACE) is a general process for capturing operator outcome data during metaheuristic search, training classifiers to predict whether an operator will yield an improved solution, and deploying those models to guide neighborhood selection during future search runs. This study introduces TRACE-VNS, a modular extension of General Variable Neighborhood Search (GVNS) applied to the Capacitated Vehicle Routing Problem (CVRP), where neighborhood selection is driven by these offline-trained models. Classifiers are trained on features extracted from GVNS traces, including action history, graph metrics, temporal state, and Upper Confidence Bound (UCB) indicators. Twelve classifiers, including tree ensembles, neural networks, and kernel-based models, are benchmarked using the Precision–Recall Area Under the Curve (PR-AUC) to evaluate predictive quality. Empirical results show that TRACE-VNS improves convergence speed and final solution quality over conventional GVNS across 84 CVRP instances. A detailed feature importance analysis identifies strong contributors, offering insights into the effective selection of operators. TRACE requires no runtime exploration or feedback loops and can generalize to other metaheuristics through minimal structural adaptation.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102058"},"PeriodicalIF":8.2,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A surrogate-assisted memetic algorithm for permutation-based combinatorial optimization problems 基于置换的组合优化问题的代理辅助模因算法

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-18 DOI: 10.1016/j.swevo.2025.102060

Takashi Ikeguchi , Kei Nishihara , Yo Kawauchi , Yuji Koguma , Masaya Nakata

{"title":"A surrogate-assisted memetic algorithm for permutation-based combinatorial optimization problems","authors":"Takashi Ikeguchi , Kei Nishihara , Yo Kawauchi , Yuji Koguma , Masaya Nakata","doi":"10.1016/j.swevo.2025.102060","DOIUrl":"10.1016/j.swevo.2025.102060","url":null,"abstract":"<div><div>Real-world applications often encounter expensive permutation-based combinatorial optimization problems (PCOPs), where solution evaluation processes become time-consuming. Although many surrogate-assisted evolutionary algorithms have been developed for expensive optimization problems, most of them are designed for expensive continuous optimization problems, not for expensive PCOPs, due to the difficulty of constructing surrogate models effective for permutation spaces. This paper presents a surrogate-assisted memetic algorithm for expensive PCOPs, designed with the following two key insights. First, Gradient Boosting Decision Tree (GBDT) regression models are adopted as surrogates tailored to discrete spaces. Because decision trees do not require distance metrics between training samples, they are well-suited to such spaces, and the boosting mechanism helps improve prediction accuracy. Additionally, we employ memetic algorithms for the search strategy to enhance both global and local search capabilities. Experiments show that the proposed method outperforms state-of-the-art algorithms for at least 41 out of all 42 PCOP instances under a limited budget of 1000 function evaluations, with improved robustness through our memetic algorithm. Furthermore, the GBDT model achieves higher prediction accuracy than other popular models, Radial Basis Function Network and Random Forest, outperforming them on more than 35 instances. These results highlight that our approach effectively enhances the synergy between surrogate models and search strategies in permutation spaces.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102060"},"PeriodicalIF":8.2,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The paradox of success in evolutionary and bioinspired optimization: Revisiting critical issues, key studies, and methodological pathways 在进化和生物启发优化成功的悖论：重访关键问题，关键研究和方法途径

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-18 DOI: 10.1016/j.swevo.2025.102063

Daniel Molina , Javier Del Ser , Javier Poyatos , Francisco Herrera

{"title":"The paradox of success in evolutionary and bioinspired optimization: Revisiting critical issues, key studies, and methodological pathways","authors":"Daniel Molina , Javier Del Ser , Javier Poyatos , Francisco Herrera","doi":"10.1016/j.swevo.2025.102063","DOIUrl":"10.1016/j.swevo.2025.102063","url":null,"abstract":"<div><div>Evolutionary and bioinspired computation are crucial for efficiently addressing complex optimization problems across diverse application domains. By mimicking processes observed in nature, like evolution itself, these algorithms offer innovative solutions beyond the reach of traditional optimization methods. They excel at finding near-optimal solutions in large, complex search spaces, making them invaluable in numerous fields. However, both areas are plagued by challenges at their core, including inadequate benchmarking, problem-specific overfitting, insufficient theoretical grounding, and superfluous proposals justified only by their biological metaphor. This overview recapitulates and analyzes in depth the criticisms concerning the lack of innovation and rigor in experimental studies within the field. To this end, we examine the judgmental positions of the existing literature in an informed attempt to guide the research community toward directions of solid contribution and advancement in these areas. We summarize guidelines for the design of evolutionary and bioinspired optimizers, the development of experimental comparisons, and the derivation of novel proposals that take a step further in the field. We provide a brief note on automating the process of creating these algorithms, which may help align metaheuristic optimization research with its primary objective (solving real-world problems), provided that our identified pathways are followed. Our conclusions underscore the need for a sustained push towards innovation and the enforcement of methodological rigor in prospective studies to fully realize the potential of these advanced computational techniques.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102063"},"PeriodicalIF":8.2,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-objective optimization for distributed flexible job shop scheduling problem with job priority 具有作业优先级的分布式柔性作业车间调度问题的多目标优化

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-18 DOI: 10.1016/j.swevo.2025.102075

Ao He, Xiahui Liu, Guiliang Gong, Zhipeng Yuan, Hongbo Huang, Yang Zhou, Jie Li

{"title":"Multi-objective optimization for distributed flexible job shop scheduling problem with job priority","authors":"Ao He, Xiahui Liu, Guiliang Gong, Zhipeng Yuan, Hongbo Huang, Yang Zhou, Jie Li","doi":"10.1016/j.swevo.2025.102075","DOIUrl":"10.1016/j.swevo.2025.102075","url":null,"abstract":"<div><div>For the distributed flexible job shop scheduling problem (DFJSP), the existing researches have predominantly focused on operation sequence, machine selection and factory assignment, and assuming that the jobs have no priority. However, in real-world manufacturing systems, production scheduling with job priority is very common and is of concern to production managers. The paper presents a DFJSP with job priority (DFJSPJP) for the first time, aiming at minimizing the makespan, total energy consumption and the weighted average time of jobs with priority. A new memetic algorithm (NMA) is designed to solve the proposed DFJSPJP. In the proposed NMA, a well-designed chromosome encoding method (CEM) is constructed to obtain a high-quality initial population. An effective local search approach (LSO) is proposed to improve the NMA’s convergence speed and fully exploit its solution space. Computational experiments conducted confirm the effectiveness of the CEM and LSO, and show that the NMA is able to easily obtain better solutions for about 90 % of the tested 60 challenging problem instances compared to other three well-known algorithms, demonstrating its superior performance on both solution quality and computational efficiency. This research will provide a theoretical basis for considering job priority issues in distributed production environments and assist manufacturers in conducting accurate production scheduling, thereby reducing resource waste and time loss caused by unreasonable production plans.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102075"},"PeriodicalIF":8.2,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A hybrid intelligent optimization algorithm for long-term production planning of open-pit mine considering carbon reduction plan 考虑碳减排计划的露天矿长期生产规划混合智能优化算法

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-18 DOI: 10.1016/j.swevo.2025.102078

Ning Li , Jinxin Liu , Liguan Wang , Bibo Dai , Shugang Zhao , Jian Chang , Haiwang Ye , Dairong Yan

{"title":"A hybrid intelligent optimization algorithm for long-term production planning of open-pit mine considering carbon reduction plan","authors":"Ning Li , Jinxin Liu , Liguan Wang , Bibo Dai , Shugang Zhao , Jian Chang , Haiwang Ye , Dairong Yan","doi":"10.1016/j.swevo.2025.102078","DOIUrl":"10.1016/j.swevo.2025.102078","url":null,"abstract":"<div><div>This study quantifies the carbon emission and its cost in the production process of open-pit mines, explores the influence of carbon emission reduction on the long-term production planning, and provides an optimal long-term production plan for open-pit mines under the background of carbon neutrality. It aims to maximize the total net present value of the mine and constructs a mathematical model for long-term production planning that integrates the carbon emission reduction plan and its associated costs. A hybrid intelligent optimization algorithm (PSBKA), based on the Particle Swarm Optimization Algorithm (PSO) and the Black-Winged Kite Optimization Algorithm (BKA), is developed. The algorithm first uses PSO to optimize the model for the primary objective and then utilizes a set of new solutions generated through a random disturbance strategy as the initial solution for BKA, performing secondary optimization on the model. An application study is conducted using a copper mine located in Arizona, USA. The results indicated that formulating and implementing a carbon emission reduction plan significantly influences long-term production planning in open-pit mining. The carbon emission reduction cost represents approximately 7 % of the mine's overall economic benefits. Compared to traditional methods, the proposed planning approach reduces the carbon emission reduction cost by $15,805 and increases the net present value by $253,811, providing an improved long-term production planning scheme for the mine.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102078"},"PeriodicalIF":8.2,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144656029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Automated algorithm selection for black-box optimization using light gradient boosting machine 基于光梯度增强机的黑盒优化自动算法选择

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-17 DOI: 10.1016/j.swevo.2025.102071

Qingbin Guo , Handing Wang , Ye Tian

{"title":"Automated algorithm selection for black-box optimization using light gradient boosting machine","authors":"Qingbin Guo , Handing Wang , Ye Tian","doi":"10.1016/j.swevo.2025.102071","DOIUrl":"10.1016/j.swevo.2025.102071","url":null,"abstract":"<div><div>Many evolutionary algorithms have been designed to address industrial black-box optimization problems in the real world. No single algorithm can outperform others across all problem instances. Algorithm selection methods aim to help users to automatically choose the best algorithm for new problems without expertise in evolutionary algorithm. However, the existing methods are implemented on a limited number of handcrafted benchmarks which lack practicality, and there is no general metric for measuring the best algorithm for black-box problems with unknown optimum. To tackle these issues, we propose an algorithm selection method for black-box optimization using light gradient boosting machine, where a tree-based random instance generation method is introduced to create diverse problem instances simulating real-world cases, and a metric is proposed to evaluate the performance of evolutionary algorithms on real-world black-box optimization considering both convergence speed and value. Experimental results show that our method achieves an accuracy of 72.23% on our generated dataset, and has lower computational cost compared to existing methods.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102071"},"PeriodicalIF":8.2,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Lightweight bi-level optimization algorithm for synchronized scheduling of production and transportation in a reconfigurable flexible job shop 可重构柔性作业车间生产与运输同步调度的轻量级双层优化算法

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-17 DOI: 10.1016/j.swevo.2025.102080

Lixin Cheng , Qiuhua Tang , Zikai Zhang , Jing Wang

{"title":"Lightweight bi-level optimization algorithm for synchronized scheduling of production and transportation in a reconfigurable flexible job shop","authors":"Lixin Cheng , Qiuhua Tang , Zikai Zhang , Jing Wang","doi":"10.1016/j.swevo.2025.102080","DOIUrl":"10.1016/j.swevo.2025.102080","url":null,"abstract":"<div><div>In reconfigurable flexible job shops, jobs have flexible processing routes and machine configurations change dynamically. This makes the transportation scheduling of Work-in-Progress (WIP) between machines highly complex and crucial to the production process. Coordinating production and transportation can significantly boost workshop efficiency. Thus, the synchronization of production and transportation scheduling is addressed. A bi-level scheduling model is developed. The upper-level minimizes production costs in production scheduling, considering flexibility and reconfigurability. The lower-level minimizes transportation costs in transportation scheduling, considering speed-adjustable Automated Guided Vehicles (AGVs). To solve this complex problem efficiently, a lightweight bi-level optimization algorithm is designed. In it, an accurate surrogate model and an improved metaheuristic are performed sequentially to achieve the lightweight evaluation and high-fidelity evaluation of the lower-level optima. Ten rules that contain problem-related knowledge are discovered by rule mining technologies including gene expression programming and Q-learning. Since these rules can better reflect problem characteristics, rule-based features are extracted to improve the accuracy of the surrogate model. Experimental results show that all discovered rules, especially the dynamic adaptive rule, are highly effective in generating high-performance solutions. The rule-based surrogate model can quickly and accurately estimate the lower-level optima. By incorporating this surrogate model, the proposed lightweight algorithm cuts down on computing budget without sacrificing accuracy, outperforming other bi-level optimization algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102080"},"PeriodicalIF":8.2,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Surrogate-assisted evolutionary algorithm with stage-adaptive infill sampling criterion for expensive multimodal multi-objective optimization 基于阶段自适应填充采样准则的代理辅助进化算法用于昂贵的多模态多目标优化

IF 8.2 1区计算机科学

Swarm and Evolutionary Computation Pub Date : 2025-07-17 DOI: 10.1016/j.swevo.2025.102068

Yufei Yang , Changsheng Zhang , Yi Liu , Haitong Zhao

{"title":"Surrogate-assisted evolutionary algorithm with stage-adaptive infill sampling criterion for expensive multimodal multi-objective optimization","authors":"Yufei Yang , Changsheng Zhang , Yi Liu , Haitong Zhao","doi":"10.1016/j.swevo.2025.102068","DOIUrl":"10.1016/j.swevo.2025.102068","url":null,"abstract":"<div><div>The key issue in handling expensive multimodal multi-objective optimization problems is to balance convergence and diversity in both the decision and objective spaces with limited function evaluations available. To tackle this issue, this paper proposes a surrogate-assisted multimodal multi-objective evolutionary algorithm with stage-adaptive infill sampling criterion. In the proposed algorithm, a multi-surrogate cooperative framework is developed, where multiple extreme gradient boosting models are used to approximate the objective functions for replacing real function evaluations, and a self-organizing map (SOM) network is used to learn the topologies of Pareto sets in the decision space and corresponding features in the objective space for reducing the approximation errors. Then, a stage-adaptive infill sampling criterion is designed to select the most suitable candidates for expensive function evaluations. Specifically, in the first stage, a convergence-first infill sampling criterion is used to accelerate convergence to the global Pareto front; In the second stage, an indicator-based infill sampling criterion according to neuron weights of the SOM network and a diversity-based infill sampling criterion are used to improve diversity in decision and objective spaces. Experimental results on two benchmark test suites demonstrate the competitiveness of the proposed algorithm against eight state-of-the-art methods.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"98 ","pages":"Article 102068"},"PeriodicalIF":8.2,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0