Article
Computer Science, Artificial Intelligence
Ming Yang, Aimin Zhou, Xiaofen Lu, Zhihua Cai, Changhe Li, Jing Guan
Summary: Cooperative co-evolution (CC) is an optimization problem decomposition strategy that reduces the difficulty of solving large-scale optimization problems. This paper presents a new CC framework called CCFR3, which adaptively evaluates the contribution of each subpopulation and improves the algorithm's performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Automation & Control Systems
Ya-Hui Jia, Yi Mei, Mengjie Zhang
Summary: This study proposes a new contribution-based cooperative co-evolutionary algorithm to address non-separable large-scale problems with overlapping subcomponents. The algorithm outperforms existing methods due to its novel decomposition method and optimization framework.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Mathematics
Aleksei Vakhnin, Evgenii Sopov, Eugene Semenkin
Summary: The study introduces a new self-adaptive approach, COSACC-LS1, which combines multiple ideas from state-of-the-art algorithms to solve LSGO problems, showing adaptive advantages and improving algorithm efficiency and effectiveness.
Article
Computer Science, Artificial Intelligence
Ming Yang, Aimin Zhou, Changhe Li, Xin Yao
Summary: Cooperative co-evolution (CC) is an efficient evolutionary framework for large-scale optimization problems, but its performance is affected by variable decomposition. To reduce computational costs, an efficient recursive differential grouping (ERDG) method is proposed in this article, which utilizes historical information to examine variable interrelationships and improve performance.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Artificial Intelligence
Yongsheng Liang, Zhigang Ren, Lin Wang, Hanqing Liu, Wenhao Du
Summary: This study proposes a surrogate-assisted cooperative signal optimization (SCSO) method to address the performance deterioration issue in large-scale traffic signal optimization problems. By decomposing the traffic network into tractable sub-networks and cooperatively optimizing them with a surrogate-assisted optimizer, SCSO effectively reduces the computational burden and improves the efficiency of signal setting.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Wu Deng, Shifan Shang, Xing Cai, Huimin Zhao, Yongquan Zhou, Huayue Chen, Wuquan Deng
Summary: The proposed HMCFQDE combines quantum evolutionary algorithm (QEA) and cooperative coevolution evolutionary algorithm (CCEA) to improve the solution efficiency and search speed. A new hybrid mutation strategy is designed to enhance convergence accuracy and stability in solving high-dimensional complex functions.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Jianxing Liu, Zhibo Sui, Xiaoxia Li, Jie Yang
Summary: The paper introduces a hybrid distributed evolutionary model to address the large scale flexible job-shop scheduling problem (LSFJSP), which decomposes the population and achieves coevolution using division layer and coevolution layer. The model demonstrates better optimization results and higher computational efficiency compared to other distributed models.
Article
Computer Science, Artificial Intelligence
Xiaoliang Ma, Zhitao Huang, Xiaodong Li, Lei Wang, Yutao Qi, Zexuan Zhu
Summary: This article introduces a merged differential grouping (MDG) method, which is a divide-and-conquer strategy to solve large-scale global optimization problems. By decomposing the problem into manageable subproblems and using binary search to group variables, the method improves the efficiency and accuracy of problem decomposition.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)
Article
Computer Science, Artificial Intelligence
Lin Li, Wei Fang, Yi Mei, Quan Wang
Summary: Cooperative coevolution is an effective strategy for solving large-scale global optimization by decomposing the problem into lower-dimensional subproblems. Differential Grouping is a competitive decomposition method, but faces challenges with overlapping problems. A novel fuzzy decomposition algorithm based on interaction degree has been proposed to address this issue.
Article
Computer Science, Information Systems
Chen Huang, Xiangbing Zhou, Xiaojuan Ran, Yi Liu, Wuquan Deng, Wu Deng
Summary: In this article, a novel co-evolutionary method called TPCSO is proposed to enhance the convergence and search ability of CSO. The modified CSO divides the population into two sub-populations and adjusts the update strategy based on diversity and convergence requirements. Experimental results show that TPCSO can effectively solve large-scale optimization problems and achieve optimal results with higher accuracy compared to other algorithms.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Jin Liu, Ruochen Liu, Xilong Zhang
Summary: This paper proposes a method for detecting the interaction of decision variables in large-scale multi-objective optimization problems. By applying recursive grouping and analyzing the contribution of each group to the problem, the proposed method optimizes the solution process. Experimental results show that it has competitive performance compared with state-of-the-art algorithms.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Ali Kelkawi, Imtiaz Ahmad, Mohammed El-Abd
Summary: The cooperative coevolution framework improves the quality and computational speed of metaheuristic algorithms for solving continuous large-scale global optimization problems by dividing the problem into subcomponents. This work proposes a distributed implementation of the framework on the Apache Spark platform, utilizing its features to enhance computational speed while maintaining comparable search quality. The proposed implementation achieves a speedup of up to x3.36 on large-scale global optimization benchmarks using Apache Spark.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)
Article
Thermodynamics
WanJun Yin, Xuan Qin
Summary: This paper proposes a method to solve the optimal scheduling problem of large-scale electric vehicles connected to the grid. By considering multiple factors and using a high-confidence wind power scenario, the collaborative optimization of coal-fired power generation, wind power generation, and electric vehicles is achieved.
Article
Computer Science, Artificial Intelligence
Ali Kelkawi, Mohammed El-Abd, Imtiaz Ahmad
Summary: In this paper, a parallel implementation of the cooperative coevolution framework for solving continuous large-scale optimization problems is proposed, utilizing GPU and CUDA platform to optimize problem subcomponents in parallel, leading to significant speedup in the optimization process.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Yuxin Han, Xingsheng Gu
Summary: This paper focuses on the large scale multi-stage multi-product batch plant scheduling problem and proposes an efficient cooperative hybrid evolutionary algorithm. Novel encoding scheme and specific operations tailored for MMSP optimization, along with a local search algorithm to further enhance efficiency.
Article
Engineering, Industrial
Xing Wan, Xingquan Zuo, Xiaodong Li, Xinchao Zhao
Summary: The multi-row facility layout problem (MRLP) is an important design problem in real life. Existing studies often overlook clearances between machines or only consider minimum clearances. This paper proposes using larger clearances between adjacent machines to achieve lower material flow cost, while optimizing layout area. By combining mixed integer programming and multi-objective greedy randomised adaptive search procedure, an effective solution is provided.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2022)
Article
Automation & Control Systems
Guanqiang Gao, Yi Mei, Ya-Hui Jia, Will N. Browne, Bin Xin
Summary: The problem of multipoint dynamic aggregation involves designing an optimal plan for multiple robots to collaboratively execute tasks while considering the changing task demands and abilities of the robots. To effectively address this challenge, a new metaheuristic algorithm called adaptive coordination ant colony optimization (ACO) is proposed, which significantly outperforms existing methods in terms of effectiveness and efficiency.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Amir H. Gandomi, Kalyanmoy Deb, Ronald C. Averill, Shahryar Rahnamayan, Mohammad Nabi Omidvar
Summary: To solve complex real-world problems, a concept-based approach called variable functioning (Fx) is introduced to reduce optimization variables and narrow down the search space. By using problem structure analysis and engineering expert knowledge, the Fx method enhances the steel frame design optimization process. Coupled with particle swarm optimization and differential evolution algorithms, the proposed approach improves the convergence rate and final design of frame structures.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2023)
Article
Computer Science, Artificial Intelligence
Mazhar Ansari Ardeh, Yi Mei, Mengjie Zhang, Xin Yao
Summary: The uncertain capacitated arc routing problem (UCARP) is a difficult combinatorial optimization problem in logistics. Genetic programming (GP) hyper-heuristic has been successfully applied to evolve routing policies for this problem. However, existing methods are not sufficient in handling the change from one instance to another. To address this issue, we propose a novel knowledge transfer GP with an auxiliary population. Experimental results show that our method outperforms the state-of-the-art GP approaches in terms of both final performance and convergence speed.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Yi Mei, Qi Chen, Andrew Lensen, Bing Xue, Mengjie Zhang
Summary: Explainable artificial intelligence (XAI) has gained significant attention, particularly in critical domains like self-driving cars, law, and healthcare. Genetic programming (GP), an evolutionary algorithm for machine learning, has been shown to produce more interpretable models compared to neural networks. This article comprehensively reviews studies on how GP can improve model interpretability, either by directly evolving interpretable models or by explaining opaque models using GP or simpler models. The survey highlights the potential of GP in addressing the tradeoff between model accuracy and interpretability.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Automation & Control Systems
Ya-Hui Jia, Yi Mei, Mengjie Zhang
Summary: This article investigates the effectiveness of using numeric representations on stochastic routing problems and uncertain capacitated arc routing problems. Linear representation, artificial neural network (ANN) representation, and tree representation are implemented and compared. Experimental results show that the tree representation is the best choice, but numeric representations, especially the ANN representation, demonstrate competitive performance in most test instances.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Automation & Control Systems
Ya-Hui Jia, Yi Mei, Mengjie Zhang
Summary: This article proposes a two-stage swarm optimizer with local search for solving the large-scale WDN optimization problem. The optimization process is divided into an exploration stage and an exploitation stage, with an improved level-based learning optimizer and two new local search algorithms. Experiments show that the proposed algorithm outperforms state-of-the-art metaheuristic algorithms.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Yuzhou Zhang, Yi Mei, Haiqi Zhang, Qinghua Cai, Haifeng Wu
Summary: In this article, a new divide-and-conquer strategy is proposed for solving Large Scale MDCARP, which introduces a restricted global optimization stage and a problem-specific Task Moving among Sub-problems process. By incorporating these into the RoCaSH algorithm, the resultant RoCaSH2 algorithm outperforms state-of-the-art algorithms in terms of efficiency and effectiveness on a wide range of instances.
IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE
(2023)
Article
Computer Science, Information Systems
Meng Xu, Yi Mei, Shiqiang Zhu, Beibei Zhang, Tian Xiang, Fangfang Zhang, Mengjie Zhang
Summary: Dynamic Workflow Scheduling in Fog Computing is a significant optimization problem that involves the coordination of cloud servers, mobile devices, and edge servers. This article proposes a new problem model and simulator, as well as a Multi-Tree Genetic Programming method to address the problem. Experimental results demonstrate that the proposed method achieves significantly better performance across all tested scenarios.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhixing Huang, Yi Mei, Fangfang Zhang, Mengjie Zhang
Summary: Dispatching rules are important for dynamic scheduling problems. This paper proposes a grammar-guided LGP method to improve the performance of LGP in job shop scheduling. The simulation results show that grammar-guided LGP has better training efficiency and can produce solutions with good explanations. It significantly improves the overall test effectiveness when the number of LGP registers increases.
PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Jordan MacLachlan, Yi Mei, Fangfang Zhang, Mengjie Zhang, Jessica Signal
Summary: The task of emergency medical response in modern municipalities is valuable. To ensure minimal response times, resource allocation and maximizing ambulance coverage can be done by human experts or automation.
PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Jiyuan Pei, Hao Tong, Jialin Liu, Yi Mei, Xin Yao
Summary: This paper proposes an empirical analysis method to study the relationship between search operators in combinatorial optimization problems. The correlation between their local optima is measured to quantify their relationship. The results show a consistent pattern in the correlation between commonly used operators. Based on this, a novel approach for adaptively selecting operators is proposed, which outperforms commonly used methods.
PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Joao Guilherme Cavalcanti Costa, Yi Mei, Mengjie Zhang
Summary: The Vehicle Routing Problem (VRP) is a complex problem with numerous applications in logistics and supply chains. Selecting the optimal VRP techniques is challenging due to various possible scenarios. This paper focuses on the initialization part of a local search-based metaheuristic and proposes using machine learning techniques to predict the effectiveness of different construction heuristics solutions. Results show that this method can help select the best or improving method for most instances, especially for large-scale VRP instances.
PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Hayden Andersen, Andrew Lensen, Will N. Browne, Yi Mei
Summary: Counterfactual explanations, a popular explainable AI technique, aim to provide contrastive answers to hypothetical questions. This work introduces two novel algorithms, Particle Swarm Optimization (PSO) and Differential Evolution (DE), to generate counterfactual explanations, without assuming anything about the underlying model or data structure. The generated explanations are sparser compared to previous related work.
2022 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC)
(2022)
Proceedings Paper
Computer Science, Cybernetics
Shaolin Wang, Yi Mei, Mengjie Zhang
Summary: This paper proposes a Local Ranking Explanation (LRE) method for explaining GP-evolved routing policies for the Uncertain Capacitated Arc Routing Problem (UCARP). Experimental results show that this method can provide more interpretable linear models to explain the routing policies in most decision situations.
PROCEEDINGS OF THE 2022 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE (GECCO'22)
(2022)