Project Articles

Special issue on evolutionary computation         当前，人工智能正处于快速变革和更新迭代的关键时期，面临基础理论薄弱、计算效率低下、算法应用受限等挑战，新型可靠高效的进化计算相关研究受到了各界广泛关注。进化计算无需预先的训练和数据积累，通过模拟生物进化过程和群体智能行为，依靠演化过程和自然选择机制，就能找到最优的解决方案，是一种知识创新和规律获取的过程。这使得进化计算具有很好的通用性，能够在探索过程中发现人类未曾知晓的知识，为解决实际应用中的一系列复杂问题提供最优的解决方案。         为此《计算机应用》组织“进化计算专题”，邀约了来自欧洲科学院院士、IEEE Fellow等多位专家及其团队的11篇优质论文，分别展示了大规模进化计算、多目标进化计算等研究热点的研究现状和最新研究成果，希望能为从事相关工作的读者提供借鉴和参考。         在多任务进化计算的前沿研究方面：IEEE Fellow、新加坡科学技术研究局人工智能首席科学家Yew-Soon Ong教授及其团队的论文“优化场景视角下的进化多任务优化综述”则以优化场景作为视角进行切入，对进化多任务优化进行了前沿研究；国家级领军人才、IEEE Fellow公茂果教授及其团队的论文“多任务优化算法及应用研究综述”对现有多任务进化计算的相关工作进行了系统性的梳理，并给出了未来研究发展的展望。         在大规模进化计算的前沿研究方面：国家杰青、科技部石油化工智能制造创新团队负责人杜文莉教授及其团队的论文“面向大规模重叠问题的两阶段差分分组方法”针对大规模重叠问题的可分性质进行了分析探索，提出了新型的差分分组方法，以对大规模问题进行高效分解；此外，杜文莉教授及其团队的另一篇高质量论文“基于多时间尺度协同的大规模原油调度进化算法”对大规模原油调度问题的特性进行分析，设计了新型的多时间尺度协同的进化算法，实现了对大规模原油调度问题的高效求解。         在多目标进化计算的前沿研究方面：欧洲科学院院士、IEEE Fellow、国家特聘专家金耀初院士及其团队的论文“GPU加速的演化算法求解多目标流水车间调度问题”通过结合GPU计算技术，提出了计算高效的演化算法，实现了对多目标流水车间调度问题的快速求解；IEEE Fellow、科技部机器智能创新团队负责人、国家杰青、长江学者、国家级领军人才张军教授及其团队的论文“概率驱动的动态多目标多智能体协同调度进化优化”针对多智能体协同调度问题的动态变化特性，提出了概率驱动型的多目标进化计算方法，实现了对多智能体协同调度的高效优化。         在约束进化计算的前沿研究方面：IEEE Fellow、长江学者唐珂教授及其团队的论文“机会约束的多选择背包问题的遗传算法求解”结合机会约束和遗传算法，实现了对多选择背包问题的高效求解；长江学者、计算智能与信号处理教育部重点实验室主任张兴义教授及其团队的论文“面向约束多目标优化的进化计算与梯度下降联合优化算法”通过结合进化计算与梯度下降方法，提出了适用于多目标约束场景的高效算法；国家重点研发科技创新2030“新一代人工智能”重大项目首席科学家、国家优青陈伟能教授及其团队的论文“分布式数据驱动的多约束进化优化算法”利用分布式的数据驱动方法，提出了高效能的多约束进化优化算法。         在进化计算的前沿应用研究方面：国家杰青、科技部科技创新领军人才段海滨教授及其团队的论文“基于有限忍耐度鸽群优化的无人机近距空战机动决策”给出了群体智能算法在无人机自动决策中的应用示范；IEEE Fellow、岭南大学副校长邝得互教授及其团队的论文“进化双层自适应局部特征选择”提出了双层自适应的进化计算方法，并在特征选择问题中进行了应用。         本专题集结了一系列进化计算领域最新的研究成果，涵盖了大规模进化计算、多目标进化计算、多任务进化计算、约束进化计算等核心方法及其应用。研究涉及优化问题、数据挖掘、机器学习、智能控制等多个领域，展现了进化计算在不同领域的广泛应用和巨大潜力。         我们希望本专题论文能够激发更多学者对进化计算相关研究的兴趣，推动该领域的发展。由于组稿时间和篇幅有限等原因，本专题无法全面覆盖进化计算领域各方面的最新研究进展，不当之处请同行学者批评指正！最后我们衷心感谢所有邀约作者和审稿专家的辛勤工作和宝贵贡献，感谢编辑部的全力支持和辛勤付出！ 张军（南开大学），黎建宇（南开大学） 2024年4月

Default Latest Most Read

Please wait a minute... For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails Select Two-stage differential grouping method for large-scale overlapping problems Maojiang TIAN, Mingke CHEN, Wei DU, Wenli DU Journal of Computer Applications    2024, 44 (5): 1348-1354.   DOI: 10.11772/j.issn.1001-9081.2024020255 Abstract （74）   HTML （11）    PDF （738KB）（45）       Knowledge map    Save Large-scale overlapping problems are prevalent in practical engineering applications， and the optimization challenge is significantly amplified due to the existence of shared variables. Decomposition-based Cooperative Co-evolution （CC） algorithms have demonstrated promising performance in addressing large-scale overlapping problems. However， certain novel CC frameworks designed for overlapping problems rely on grouping methods for the identification of overlapping problem structures and the current grouping methods for large-scale overlapping problems fail to consider both accuracy and efficiency simultaneously. To address the above problems， a Two-Stage Differential Grouping （TSDG） method for large-scale overlapping problems was proposed， which achieves accurate grouping while significantly reducing computational resource consumption. In the first stage， a grouping method based on the finite difference principle was employed to efficiently identify all subcomponents and shared variables. To enhance both stability and accuracy in grouping， a grouping refinement method was proposed in the second stage to examine the information of the subcomponents and shared variables obtained in the previous stage and correct inaccurate grouping results. Based on the synergy of the two stages， TSDG achieves efficient and accurate decomposition of large-scale overlapping problems. Extensive experimental results demonstrate that TSDG is capable of accurately grouping large-scale overlapping problems while consuming fewer computational resources. In the optimization experiment， TSDG exhibits superior performance compared to state-of-the-art algorithms for large-scale overlapping problems. Table and Figures | Reference | Related Articles | Metrics Select Multi-timescale cooperative evolutionary algorithm for large-scale crude oil scheduling Wanting ZHANG, Wenli DU, Wei DU Journal of Computer Applications    2024, 44 (5): 1355-1363.   DOI: 10.11772/j.issn.1001-9081.2024020254 Abstract （56）   HTML （14）    PDF （2180KB）（41）       Knowledge map    Save Aiming to solve the problems of large-scale resources， complex constraints， and difficult cooperation of multi-timescale decision-making in the crude oil scheduling process， a Multi-Timescale Cooperation Evolutionary Algorithm （MTCEA） was proposed. Firstly， a large-scale multi-timescale crude oil scheduling optimization model was established according to the scale structure and actual demand of oil refining enterprises， which consists of a resource-oriented medium- and long-term scheduling model and an operation-oriented short-term scheduling model， and achieves a reasonable allocation of crude oil resources through employing a dynamic grouping strategy of crude oil resources to satisfy the requirements of different scheduling scales， multi-timescale characteristics， and fine production. Secondly， to promote the integration of scheduling decisions at different time scales， an evolutionary algorithm based on multi-timescale cooperation was designed and solved by constructing subproblems for the continuous decision variables in scheduling models at different time scales to achieve cooperation optimization between scheduling decisions at different time scales. Finally， MTCEA was verified in three practical industrial cases. Compared with three representative large-scale evolutionary optimization algorithms （i.e.， Competitive Swarm Optimizer （CSO）， Self-adaptive Differential Evolution with Modified Multi-Trajectory Search （SaDE-MMTS）， and Mixture Model-based Evolution Strategy （MMES）） and three high-performance Mixed Integer Non-Linear Programming （MINLP） mathematical solvers （ANTIGONE （Algorithms for coNTinuous/Integer Global Optimization of Nonlinear Equations）， SCIP （Solving Constraint Integer Programs）， and SHOT （Supporting Hyperplane Optimization Toolkit））， the results show that the metrics of the solution optimality and stability of MTCEA are improved by more than 30% and 25%， respectively. These significant performance improvements demonstrate the practical application value and advantages of MTCEA in large-scale multi-timescale crude oil scheduling decisions. Table and Figures | Reference | Related Articles | Metrics Select Research review of multitasking optimization algorithms and applications Yue WU, Hangqi DING, Hao HE, Shunjie BI, Jun JIANG, Maoguo GONG, Qiguang MIAO, Wenping MA Journal of Computer Applications    2024, 44 (5): 1338-1347.   DOI: 10.11772/j.issn.1001-9081.2024020209 Abstract （102）   HTML （11）    PDF （1486KB）（81）       Knowledge map    Save Evolutionary MultiTasking Optimization （EMTO） is one of the new methods in evolutionary computing， which can simultaneously solve multiple related optimization tasks and enhance the optimization of each task through knowledge transfer between tasks. In recent years， more and more research on evolutionary multitasking optimization has been devoted to utilizing its powerful parallel search capability and potential for reducing computational costs to optimize various problems， and EMTO has been used in a variety of real-world scenarios. The researches and applications of EMTO were discussed from four aspects： principle， core design， applications， and challenges. Firstly， the general classification of EMTO was introduced from two levels and four aspects， including single-population multitasking， multi-population multitasking， auxiliary task， and multiform task. Next， the core component design of EMTO was introduced， including task construction and knowledge transfer. Finally， its various application scenarios were introduced and a summary and outlook for future research was provided. Table and Figures | Reference | Related Articles | Metrics Select Review of evolutionary multitasking from the perspective of optimization scenarios Jiawei ZHAO, Xuefeng CHEN, Liang FENG, Yaqing HOU, Zexuan ZHU, Yew‑Soon Ong Journal of Computer Applications    2024, 44 (5): 1325-1337.   DOI: 10.11772/j.issn.1001-9081.2024020208 Abstract （127）   HTML （18）    PDF （1383KB）（122）       Knowledge map    Save Due to the escalating complexity of optimization problems， traditional evolutionary algorithms increasingly struggle with high computational costs and limited adaptability. Evolutionary MultiTasking Optimization （EMTO） algorithms have emerged as a novel solution， leveraging knowledge transfer to tackle multiple optimization issues concurrently， thereby enhancing evolutionary algorithms’ efficiency in complex scenarios. The current progression of evolutionary multitasking optimization research was summarized， and different research perspectives were explored by reviewing existing literature and highlighting the notable absence of optimization scenario analysis. By focusing on the application scenarios of optimization problems， the scenarios suitable for evolutionary multitasking optimization and their fundamental solution strategies were systematically outlined. This study thus could aid researchers in selecting the appropriate methods based on specific application needs. Moreover， an in-depth discussion on the current challenges and future directions of EMTO were also presented to provide guidance and insights for advancing research in this field. Table and Figures | Reference | Related Articles | Metrics Select Novel genetic algorithm for solving chance-constrained multiple-choice Knapsack problems Xuanfeng LI, Shengcai LIU, Ke TANG Journal of Computer Applications    2024, 44 (5): 1378-1385.   DOI: 10.11772/j.issn.1001-9081.2024010113 Abstract （47）   HTML （4）    PDF （1793KB）（23）       Knowledge map    Save Chance-Constrained Multi-Choice Knapsack Problem （CCMCKP） is a class of NP-hard combinatorial optimization problems with important practical applications. However， there is a lack of research on the solution methods for this problem. The first framework for solving CCMCKP was proposed for this problem， and two solution methods were established based on this framework， including the dynamic programming-based method RA-DP and the genetic algorithm-based method RA-IGA. RA-DP is an exact method with optimality guarantee， but it can only solve small-scale problem instances within a time budget of 1 hour. In contrast， RA-IGA is an approximation method with better scalability. Simulation experimental results verify the performance of the proposed methods. On small-scale problem instances， both RA-DP and RA-IGA can find the optimal solutions. On the medium- and large-scale problem instances， RA-IGA exhibits significantly higher efficiency than RA-DP， always obtaining feasible solutions quickly within 1 hour. In future research on CCMCKP， RA-DP and RA-IGA can be considered as baseline methods， and the benchmark set considered in this work can also be used as a standard benchmark test set. Table and Figures | Reference | Related Articles | Metrics Select GPU-accelerated evolutionary optimization of multi-objective flow shop scheduling problems Tao JIANG, Zhenyu LIANG, Ran CHENG, Yaochu JIN Journal of Computer Applications    2024, 44 (5): 1364-1371.   DOI: 10.11772/j.issn.1001-9081.2024010028 Abstract （79）   HTML （10）    PDF （1464KB）（40）       Knowledge map    Save In the realms of intelligent manufacturing and environmental sustainability， the significance of multi-objective scheduling in orchestrating a balance among production efficiency， cost management， and environmental conservation is paramount. Contemporary research indicates that CPU-based scheduling solutions are constrained by suboptimal efficiency and responsiveness， particularly when managing tasks of considerable scale. Consequently， the parallel computational prowess of GPUs heralds a novel avenue for the refinement of extensive flow shop scheduling challenges. For the multi-objective No-Wait Flow shop Scheduling Problem （NWFSP）， with the concurrent objectives of minimizing both the makespan and the Total Energy Consumption （TEC）， a Mixed-Integer Linear Programming model （MILP） was formulated to delineate the problem， and a bespoke GPU-accelerated tensorized evolutionary algorithm named Tensor-GPU-NSGA-Ⅱ was introduced for problem-solving. The ingenuity of Tensor-GPU-NSGA-Ⅱ resides in its tensorized algorithm for the computation of the makespan and TEC within the NWFSP framework， as well as in converting the conventional CPU-based serial individual updating to a GPU-driven parallel population renewal process. Empirical results demonstrate that for a scenario involving 500 jobs and 20 machines， Tensor-GPU-NSGA-Ⅱ realizes an enhancement in computational efficiency by a speedup of 9 761.75 over the traditional NSGA-Ⅱ algorithm. Furthermore， this acceleration efficacy markedly surges as the population scale expands. Table and Figures | Reference | Related Articles | Metrics Select Probability-driven dynamic multiobjective evolutionary optimization for multi-agent cooperative scheduling Xiaofang LIU, Jun ZHANG Journal of Computer Applications    2024, 44 (5): 1372-1377.   DOI: 10.11772/j.issn.1001-9081.2023121865 Abstract （42）   HTML （5）    PDF （1353KB）（20）       Knowledge map    Save In multi-agent systems， there are multiple cooperative tasks that change with time and multiple conflict optimization objective functions. To build a multi-agent system， the dynamic multiobjective multi-agent cooperative scheduling problem becomes one of critical problems. To solve this problem， a probability-driven dynamic prediction strategy was proposed to utilize the probability distributions in historical environments to predict the ones in new environments， thus generating new solutions and realizing the fast response to environmental changes. In detail， an element-based representation for probability distributions was designed to represent the adaptability of elements in dynamic environments， and the probability distributions were gradually updated towards real distributions according to the best solutions found by optimization algorithms in each iteration. Taking into account continuity and relevance of environmental changes， a fusion-based prediction mechanism was built to predict the probability distributions and to provide a priori knowledge of new environments by fusing historical probability distributions when the environment changes. A new heuristic-based sampling mechanism was also proposed by combining probability distributions and heuristic information to generate new solutions for updating out-of-date populations. The proposed probability-driven dynamic prediction strategy can be inserted into any multiobjective evolutionary algorithms， resulting in probability-driven dynamic multiobjective evolutionary algorithms. Experimental results on 10 dynamic multiobjective multi-agent cooperative scheduling problem instances show that the proposed algorithms outperform the competing algorithms in terms of solution optimality and diversity， and the proposed probability-driven dynamic prediction strategy can improve the performance of multiobjective evolutionary algorithms in dynamic environments. Table and Figures | Reference | Related Articles | Metrics Select Short-range UAV air combat maneuver decision-making via finite tolerance pigeon-inspired optimization Zhiqiang ZHENG, Haibin DUAN Journal of Computer Applications    2024, 44 (5): 1401-1407.   DOI: 10.11772/j.issn.1001-9081.2023121837 Abstract （45）   HTML （5）    PDF （2642KB）（17）       Knowledge map    Save Due to the rapid change of the situation during the confrontation， the autonomous maneuver decision-making for short-range Unmanned Aerial Vehicle （UAV） air combat is difficult and complex， which is a difficult point in air combat. To address this issue， a short-range UAV air combat maneuver decision-making method based on Finite Tolerance Pigeon-Inspired Optimization （FTPIO） algorithm was proposed. Two parts were designed in the proposed method： opponent action prediction based on the maneuver library and optimization solution of maneuver control and execution time based on FTPIO algorithm. To improve the global exploration ability of the basic Pigeon-Inspired Optimization （PIO） algorithm， the finite tolerance strategy was introduced. When the pigeon individual failed to find a better solution in several iterations， its attributes would been reset to avoid falling into the local optimal trap. The optimization variables used in the proposed method were the increments of the control variables of UAV motion model， which broke the limitations of the maneuver library. Simulation and adversarial testing results with the MiniMax method， basic PIO algorithm， and Particle Swarm Optimization （PSO） algorithm show that the proposed maneuver decision-making method can effectively defeat opponents during confrontation and generate more flexible deceptive maneuver behaviors. Table and Figures | Reference | Related Articles | Metrics Select Evolutionary bi-level adaptive local feature selection Lin GAO, Yu ZHOU, Tak Wu KWONG Journal of Computer Applications    2024, 44 (5): 1408-1414.   DOI: 10.11772/j.issn.1001-9081.2023121829 Abstract （63）   HTML （2）    PDF （2984KB）（17）       Knowledge map    Save Local Feature Selection （LFS） methods partition the sample space into multiple local regions and select the optimal feature subset for each region to reflect local heterogeneous information. However， the existing LFS methods partition local regions around each sample and find the optimal feature subset， resulting in low optimization efficiency and limited applicability. To address this issue， a new evolutionary Bi-level adaptive Local Feature Selection （BiLFS） algorithm was proposed. The LFS problem was formulated as a bi-level optimization problem， with feature subsets and locally optimized regions as the decision variables. At the upper level， Non-dominated Sorting Genetic Algorithm Ⅱ was employed to find the optimal feature subsets for the selected local regions， with region purity and selected feature ratio as the objective functions. At the lower level， based on the upper-level solution， local region clustering analysis was used to obtain center samples within each region， followed by local region fusion to eliminate unnecessary regions and update the population of necessary regions. Experimental results on 11 UCI datasets demonstrate that BiLFS achieves an average classification accuracy up to 98.48%， and an average computation time down to 9.51% compare to those of non-adaptive LFS methods based on evolutionary algorithms， significantly improving computational efficiency to the level of linear programming-based LFS methods. Visual analysis of the locally optimized regions selected by the BiLFS algorithm during the iteration process indicates the stability and reliability of selecting necessary local regions. Table and Figures | Reference | Related Articles | Metrics Select Distributed data-driven evolutionary computation for multi-constrained optimization Fengfeng WEI, Weineng CHEN Journal of Computer Applications    2024, 44 (5): 1393-1400.   DOI: 10.11772/j.issn.1001-9081.2023121814 Abstract （39）   HTML （1）    PDF （1005KB）（21）       Knowledge map    Save Distributed data acquisition and processing in ubiquitous computing mode have brought the demand for distributed data-driven optimization. To address the challenges such as distributed data acquisition， asynchronous constraints evaluation and incomplete information， a Distributed Data-Driven Evolutionary Algorithm （DDDEA） framework for multi-constrained optimization was constructed. A series of terminal nodes were responsible for data provision and distributed evaluation， while the server nodes were responsible for global evolutionary optimization. Based on this framework， a specific algorithm instance was implemented， the terminal nodes utilized their local data to construct a Radial Basis Function （RBF） model to assist the differential evolution of the server node. Experimental comparison with three centralized data-driven evolutionary algorithms for multi-constrained optimization on two standard test suites show that， the DDDEA achieves significant optimal results in 68.4% of test cases and has a success rate of 1.00 in finding feasible solutions in 84.2% of test cases. Therefore， the DDDEA has satisfactory global search and convergence abilities. Table and Figures | Reference | Related Articles | Metrics Select Hybrid optimizer combining evolutionary computation and gradient descent for constrained multi-objective optimization Ye TIAN, Jinjin CHEN, Xingyi ZHANG Journal of Computer Applications    2024, 44 (5): 1386-1392.   DOI: 10.11772/j.issn.1001-9081.2023121798 Abstract （46）   HTML （4）    PDF （1501KB）（21）       Knowledge map    Save Constrained Multi-Objective Evolutionary Algorithms （CMOEAs） are metaheuristics tailored for solving constrained multi-objective optimization problems. These algorithms use population-based stochastic search paradigms， striking balance between objectives and constrains on various optimization problems. However， they do not take advantage of gradient information of the functions， exhibiting slow convergence speed on complex problems. Nevertheless， the introduction of gradients is not trivial， as the calculation of the gradients of all the objectives and constraints are computationally expensive， and the conflicts between objectives and constraints make it difficult to determine the gradient directions. Therefore， an optimization algorithm combining evolutionary computation and Gradient Descent （GD） was proposed， namely CMOEA with Multiple Stages assisted by Gradients （CMOEA-MSG）. It consists of two stages： at the first stage， helper problems were constructed and either the gradients of objectives or the gradients of constraints were calculated， which were used to update solutions and drive the population to quickly converge towards feasible regions； at the second stage， the constraint-first principle was utilized to solve the original problem， so as to ensure the feasibility and diversity of the population. Compared with peer algorithms on LIR-CMOP， MW and DAS-CMOP test sets， CMOEA-MSG is verified to be more effective for solving constrained multi-objective optimization problems. Table and Figures | Reference | Related Articles | Metrics