一种基于最小生成树的多目标进化算法

怎样保证朝Pareto最优解的方向搜索和如何获得均匀分布且范围广泛的非支配解是多目标进化算法(MOEA)设计时的两个关键问题,它们很大程度上取决于适应度赋值和外部种群维护这两个重要部分.提出了一种基于最小生成树的多目标进化算法(MST_MOEA).在考虑了个体间支配关系的基础上,利用个体与非支配集的距离和不同等级个体的树聚集密度来对适应度赋值;在外部种群的非支配解个数超过规定的种群规模时,用树的度数和树聚集密度对其进行修剪.将其应用于不同维数下9个测试函数,并与NSGA-II,SPEA2进行对比,结果证实了算法良好的收敛性和分布性.     

基于空间距离的多目标进化算法

为提高多目标进化算法的收敛性,提出一种基于空间距离的多目标进化算法.定义一种密度估计指标--树聚集距离,在考虑非支配前沿的同时,利用个体的空间距离及树聚集距离进行个体选择操作.另外,在外部种群的非支配解个数超过规定的种群规模时,用基于个体邻近距离的维护方法对其进行维护.通过6个测试问题和5个方面的测试标准,与NSGA-Ⅱ和SPEA2进行比较,该算法在拥有更好收敛度的同时,保持良好的均匀性和分布广度.     

一种基于邻域的多目标进化算法

种群维护是多目标进化算法的重要组成部分。针对维护方法和运行效率的矛盾,提出一种基于邻域的多目标进化算法（NMOEA）。定义了一个反映个体之间邻近程度的指标--邻域包含关系,利用此关系对个体进行分布适应度分级的赋值,并用动态方法快速地对种群进行维护。通过7个测试问题和3个方面的测试标准,结果表明新算法在较快速地接近真实的最优面的同时,拥有良好的分布性。     

一种快速的基于占优树的多目标进化算法

为了解决多目标进化算法中适应值指派(fitness assignment)的耗时问题,提出了一种新颖的适应值指派方法--占优树.占优树保存了个体之间的必要信息,暗含了个体的密度信息,而且显著减少了个体之间的比较.此外,基于占优树的淘汰策略没有花费额外的代价就保存了种群多样性.在此基础上,提出了一种新的基于占优树的多目标进化算法.通过6个测试问题和3个方面的测试标准,新算法在接近真实的最优前沿和保持种群的多样性方面,与SPEA2和NSGA-II性能相当,但速度要比它们快得多.     

二维目标下分布性与收敛性结合的种群维护策略

种群维护是多目标进化算法的重要组成部分。针对传统方法在维护过程中只考虑分布性的情况,提出一种分布性与收敛性结合的种群维护策略,该方法用一种邻近个体间的相对趋近关系来表示其适应值,弥补了单纯Pareto支配关系的“粗糙性”,并用一种可调邻域的方法对种群的密集程度进行控制。将其与NSGA-II和SPEA2进行对比,实验结果表明该算法在有效保持种群分布性的同时,拥有良好的收敛性和速度。     

一种基于相似个体的多目标进化算法

分布性保持是多目标进化算法研究的一个重要方面,一个好的分布性能给决策者提供更多合理有效的选择。Pareto最优解的分布性主要体现在分布广度与均匀性两个方面。提出一种基于相似个体的多目标进化算法（SMOEA）。在种群维护中删除相似程度最大的个体;在进化操作中,选取了相似程度最大的个体进行进化。与目前经典算法NSGA-II和ε-MOEA进行比较,结果表明新算法拥有良好的分布性,同时也较好的改善了收敛性。     

具有多形态种群协同进化的多目标优化算法

为提高进化多目标优化算法在维持最优解多样性方面的性能,获得分布更均匀的Pareto非支配解集,文中提出一种具有多形态种群协同进化的多目标优化算法.该算法构建一种多形态种群协同进化架构,通过引入最小向量夹角的相似性度量方法,给出次优非支配个体选择策略,从而提高种群的多样性.算法还提出一种基于排序链表的拥挤个体删除策略,进一步提高解集分布的均匀性和宽广性.与经典算法对比结果表明,文中算法在解的分布性和多样性方面均有较好表现,尤其在解集分布均匀性方面优势较明显.     

一种基于Hypervolume指标的自适应邻域多目标进化算法

通过定义反映个体之间邻近程度的指标(个体的树邻域包含关系),在考虑个体间支配关系的基础上,利用个体与其周边个体的树邻域密度进行适应度赋值;提出了一种2,3维情况下个体独立支配区域的Hypervolume指标的计算方法,该方法用于评价个体对群体的贡献时只需要1次计算(同类方法需要2次计算);当外部种群中非支配个体数目超过规定规模时,根据个体独立支配区域的Hypervolume指标的大小对其进行修剪;在此基础上,提出了一种基于Hypervolume指标的自适应邻域多目标进化算法ANMOEA?HI.对比实验结果表明,ANMOEA?HI在保证了解集收敛性的同时亦拥有良好的分布性.     

基于最小生成树的分割区域密度聚类算法

针对传统密度聚类算法因使用全局变量导致对不平衡数据集的适应能力较差的问题,提出了一种基于最小生成树的密度聚类算法.首先进行数据集密度峰值计算,用于估计全局密度;然后通过密度聚类将数据集划分为高密度区域和低密度区域;接着构建和分割最小生成树对低密度区域内样本进行关联挖掘,用于将高密度区域与低密度区域互联;最后计算簇密度并以此作为特征进行簇合并,得到聚类结果.该算法结合图论知识,将数据按密度特征进行分块后合并处理,克服了传统密度聚类算法存在的局限性.通过选取多个不平衡人工数据集和UCI数据集对该算法进行测试,验证了该算法的有效性与鲁棒性.     

一种基于个体密度估算的多目标优化演化算法

通过在目标空间中利用目标本身信息估算个体k最近邻距离之和,作为个体的密度信息,根据个体的密度信息对群体中过剩的非劣解进行逐个去除,以便更好地维护解的多样性,由此给出了一种基于个体密度估算的多目标优化演化算法IDEMOEA。用这个算法对几个典型的多目标优化函数进行测试。测试结果表明,算法IDEMOEA求解多目标优化问题是行之有效的。     

基于最小生成树的图像分割

基于最小生成树思想,给出了一种利用改进的最小生成树进行图像分割的方案,减少了最小生成树的构建过程,对初分割的结果利用NNG算法进行合并。该方案节约了分割时间,并且对分割后的图像进行了有效的合并,达到了较好的分割效果。     

基于改进的遗传算法的多目标优化问题研究

研究多目标优化算法问题,针对传统的多目标优化算法由于计算复杂度非常高,难以获得令人满意的解等问题,在图论和遗传算法基础上,提出了一种改进的遗传算法求解多目标优化方法。首先采用二进制编码表示最小树问题,然后采用深度优先搜索算法进行图的连通性判断,给出了一种新的适应度函数,以提高算法执行速度和进化效率。最后仿真结果表明,与经典的Prim算法和Kruskal算法相比,新算法复杂度较低,并能在第一次遗传进化过程中获得一批最小生成树,适合于解决不同类型的多目标最小树问题。     

Degree constrained minimum spanning tree problem: a learning automata approach

Degree-constrained minimum spanning tree problem is an NP-hard bicriteria combinatorial optimization problem seeking for the minimum weight spanning tree subject to an additional degree constraint on graph vertices. Due to the NP-hardness of the problem, heuristics are more promising approaches to find a near optimal solution in a reasonable time. This paper proposes a decentralized learning automata-based heuristic called LACT for approximating the DCMST problem. LACT is an iterative algorithm, and at each iteration a degree-constrained spanning tree is randomly constructed. Each vertex selects one of its incident edges and rewards it if its weight is not greater than the minimum weight seen so far and penalizes it otherwise. Therefore, the vertices learn how to locally connect them to the degree-constrained spanning tree through the minimum weight edge subject to the degree constraint. Based on the martingale theorem, the convergence of the proposed algorithm to the optimal solution is proved. Several simulation experiments are performed to examine the performance of the proposed algorithm on well-known Euclidean and non-Euclidean hard-to-solve problem instances. The obtained results are compared with those of best-known algorithms in terms of the solution quality and running time. From the results, it is observed that the proposed algorithm significantly outperforms the existing method.     

A learning automata-based heuristic algorithm for solving the minimum spanning tree problem in stochastic graphs

During the last decades, a host of efficient algorithms have been developed for solving the minimum spanning tree problem in deterministic graphs, where the weight associated with the graph edges is assumed to be fixed. Though it is clear that the edge weight varies with time in realistic applications and such an assumption is wrong, finding the minimum spanning tree of a stochastic graph has not received the attention it merits. This is due to the fact that the minimum spanning tree problem becomes incredibly hard to solve when the edge weight is assumed to be a random variable. This becomes more difficult if we assume that the probability distribution function of the edge weight is unknown. In this paper, we propose a learning automata-based heuristic algorithm to solve the minimum spanning tree problem in stochastic graphs wherein the probability distribution function of the edge weight is unknown. The proposed algorithm taking advantage of learning automata determines the edges that must be sampled at each stage. As the presented algorithm proceeds, the sampling process is concentrated on the edges that constitute the spanning tree with the minimum expected weight. The proposed learning automata-based sampling method decreases the number of samples that need to be taken from the graph by reducing the rate of unnecessary samples. Experimental results show the superiority of the proposed algorithm over the well-known existing methods both in terms of the number of samples and the running time of algorithm.     

基于交叉模型的改进遗传算法

提出一种解决早熟收敛问题的改进遗传算法. 通过最小生成树聚类将种群划分为若干个子种群, 子种群内的个体之间及不同子种群间的个体之间同时进行遗传操作. 同子种群间个体的遗传操作可以保证算法的进化方向和收敛速度, 不同子种群间个体的遗传操作可以避免近亲繁殖, 提供多样性. 分别采用二进制和实数编码, 在经典的 23 个基准函数上的对比测试结果表明, 所提出算法具有较好的收敛速度和寻优能力.

     

一种具有时间约束的分布式软件可靠性评估方法

在对分布式软件的可靠性进行评估时，通常采用以分布式程序为基础建立的相应模型，然而这些模型侧重于分布式软件的可靠性，很少反映具有时间约束特征的可靠性特性，以分布式软件的可靠性模型为基础，将精简和广度优先两种方法有机结合，对文件生成树中的数据文件传输时间进行评测，确定文件生成树的状态，从而提出一种具有实时特征的分布式软件可靠性评估方法。     

Constructing Euclidean minimum spanning trees and all nearestneighbors on reconfigurable meshes

A reconfigurable mesh, R-mesh for short, is a two-dimensional array of processors connected by a grid-shaped reconfigurable bus system. Each processor has four I/O ports that can be locally connected during execution of algorithms. This paper considers the d-dimensional Euclidean minimum spanning tree (EMST) and the all nearest neighbors (ANN) problem. Two results are reported. First, we show that a minimum spanning tree of n points in a fixed d-dimensional space can be constructed in O(1) time on a √(n³)×√(n³) R-mesh. Second, all nearest neighbors of n points in a fixed d-dimensional space can be constructed in O(1) time on an n×n R-mesh. There is no previous O(1) time algorithm for the EMST problem; ours is the first such algorithm. A previous R-mesh algorithm exists for the two-dimensional ANN problem; we extend it to any d-dimensional space. Both of the proposed algorithms have a time complexity independent of n but growing with d. The time complexity is O(1) if d is a constant     

Leaf constrained minimal spanning trees solved by modified quantum-behaved particle swarm optimization

Given an undirected, connected, weighted graph, the leaf-constrained minimum spanning tree (LCMST) problem seeks a spanning tree of minimum weight among all the spanning trees of the graph with at least l leaves. In this paper, we have proposed an approach based on Quantum-Behaved Particle Swarm Optimization (QPSO) for the LCMST problem. Particle swarm optimization (PSO) is a well-known population-based swarm intelligence algorithm. Quantum-behaved particle swarm optimization (QPSO) is also proposed by combining the classical PSO philosophy and quantum mechanics to improve performance of PSO. In this paper QPSO has been modified by adding a leaping behavior. When the modified QPSO (MQPSO), falls in to the local optimum, MPSO runs a leaping behavior to leap out the local optimum. We have compared the performance of the proposed method with ML, SCGA, ACO-LCMST, TS-LCMST and ABC-LCMST, which are reported in the literature. Computational results demonstrate the superiority of the MQPSO approach over all the other approaches. The MQPSO approach obtained better quality solutions in shorter time.     

A swarm intelligence approach to the quadratic minimum spanning tree problem

The quadratic minimum spanning tree problem (Q-MST) is an extension of the minimum spanning tree problem (MST). In Q-MST, in addition to edge costs, costs are also associated with ordered pairs of distinct edges and one has to find a spanning tree that minimizes the sumtotal of the costs of individual edges present in the spanning tree and the costs of the ordered pairs containing only edges present in the spanning tree. Though MST can be solved in polynomial time, Q-MST is NP-Hard. In this paper we present an artificial bee colony (ABC) algorithm to solve Q-MST. The ABC algorithm is a new swarm intelligence approach inspired by intelligent foraging behavior of honey bees. Computational results show the effectiveness of our approach.     

求解多目标最小生成树的改进多目标蚁群算法

多目标最小生成树问题是典型的NP问题。针对此问题,提出一种改进的多目标蚁群算法。为获得更好的非劣前端,通过合理选取多个信息素扩散源与扩散策略来避免其早熟收敛,并引入非支配排序算子,提高种群多样性并避免算法过早陷入局部最优解。对比实验结果表明：对于多目标最小生成树问题,该算法是有效的,不但在求解效率和解的质量方面优于相关算法,而且随着问题规模的扩大,算法仍保持较好的性能。