基于分布估计的离散差分骨干粒子群优化

粒子群优化（PSO）和差分演化（DE）是两种新兴的优化技术,已经成功地应用于连续优化问题,但是它们至今尚不能像解决连续优化问题那样有效地处理组合优化问题。最近,有人提出差分骨干PSO（DBPSO）用于解决连续优化问题。首先提出离散DBPSO用于组合优化问题,然后在离散DBPSO中引入分布估计算法（EDA）来提高性能,把EDA抽样得到的全局统计信息和DBPSO获得的局部演化信息相结合来产生新解,形成基于EDA的离散DBPSO。实验结果表明EDA能大大提高离散DBPSO的性能。     

一种具有混合编码的二进制差分演化算法

差分演化(DE)是Storn和Price于1997年提出的一种基于个体差异重组思想的演化算法,非常适用于求解连续域上的最优化问题.首先引入"差异算子"等概念,给出DE的一种简洁算法描述,并分析了它所具有的特性.然后,为了使DE能够求解离散域上的最优化问题,基于数学变换思想引入"辅助搜索空间"和"个体混合编码"等概念,通过定义一个特殊的满射变换,在辅助搜索空间的作用下将连续域上的高效差分演化搜索变换为离散域上的同步演化搜索,由此提出了第1个二进制差分演化算法:具有混合编码的二进制差分演化算法(HBDE).接着,给出了HBDE的依概率收敛和完全收敛的定义,并利用离散Markov随机理论证明了HBDE是完全收敛的. HBDE不仅完全具有DE的各种特性和所有优点,而且非常适用于求解离散域上的最优化问题,对随机生成的大规模3-SAT问题实例和典型0/1背包问题实例的数值计算表明:该算法具有很好的全局收敛性和稳定性,其性能远远超过二进制粒子群优化算法和遗传算法.     

二进制编码差异演化算法在Agent联盟形成中的应用

在多Agent系统中,通过形成联盟可以提高Agent求解问题的能力,因此,联盟是多Agent系统的重要合作方法.从本质上讲,Agent联盟的形成是一个复杂的组合优化问题.引入差异演化算法来解决这一问题.差异演化是一种基于群体差异的演化算法,适合于求解连续空间的最优化问题.首次将以实数编码的差异演化算法应用于Agent联盟问题,提出二进制编码的差异演化算法解决组合优化问题,通过引入S型函数把变异操作的结果限制在集合{0,1}上,可以快速、高效地找出合适的Agent联盟.与遗传算法和蚁群算法的对比实验表明,该算法是正确、有效、可行的,在运行时间和解的性能上都优于相关算法.     

基于演化深度强化学习的符号网络影响最大化研究

近年来,随着互联网信息传播以及新型冠状病毒COVID-19传播链阻断等重大应用问题的出现,社会网络影响最大化问题的研究受到了科学界广泛关注.影响最大化问题旨在根据特定应用问题的传播模型,识别出最优影响种子节点集,最大化其信息传播影响.现有影响最大化算法主要针对单连接影响传播模型,将影响最大化问题模拟为离散的影响力种子节点组合选取优化问题.然而,这些算法具有较高的计算时间复杂度,且无法解决具有大规模冲突关系的符号网络影响最大化问题.针对上述问题,首先,构建适用于符号网络的正负影响传播模型以及影响最大化优化模型.其次,通过引入由神经网络构成的deep Q network来选取种子节点集,将离散的种子节点组合选取问题转化为更易优化的网络权重连续优化问题.最后,提出基于演化深度强化学习的符号网络影响最大化算法SEDRL-IM.该算法将演化算法的个体视作策略,结合演化算法的无梯度全局搜索以及强化学习的局部搜索特性,实现对deep Q network权重优化问题解的有效搜索,从而找到最优影响种子节点集.在基准符号网络以及真实社交网络数据集上的大量实验结果表明,所提算法在影响传播范围与求解效率上都优...     

基于编码转换的离散演化算法设计与应用

为了利用演化算法求解离散域上的组合优化问题,借鉴遗传算法（GA）、二进制粒子群优化（BPSO）和二进制差分演化（HBDE）中的映射方法,提出了一种基于映射变换思想设计离散演化算法的实用方法——编码转换法（ETM）,并利用一个简单有效的编码转化函数给出了求解组合优化问题的离散演化算法一般算法框架A-DisEA.为了说明ETM的实用性与有效性,首先基于A-DisEA给出了一个离散粒子群优化算法（DisPSO）,然后分别利用BPSO、HBDE和DisPSO等求解集合联盟背包问题和折扣{0-1}背包问题,通过对计算结果的比较表明：BPSO、HBDE和DisPSO的求解性能均优于GA,这不仅说明基于ETM的离散演化算法在求解KP问题方面具有良好的性能,同时也说明利用ETM方法设计离散演化算法是一种简单且有效的实用方法.     

求解RCPSP问题的带分布估计的差异演化算法

提出一种带分布估计的差异演化算法（DEED）用于求解资源受限项目调度问题（RCPSP）。该算法基于差异演化（DE）算法,利用分布估计算法（EDA）能够获得问题解空间的全局信息以及变量间的相互联系,以指导算法搜索过程,并对最优解的分布进行预测。DEED算法充分利用DE收敛速度快和EDA全局搜索优点。经标准问题库（PSPLIB）的单模式问题集验证,并与当前流行的算法进行比较,表明了DEED算法的有效性。     

多变异策略的自适应差分演化算法

差分演化(Differential Evolution,DE)算法的性能依赖于变异策略的选择和控制参数的设置.不同问题对DE的变异策略和参数的设置各不相同.为了提高DE的性能,提出一种多变异策略的自适应差分演化算法,建立由多种变异策略组成的策略池,两个主要参数自适应策略控制.为了验证所提算法的性能,在测试数据集CEC2013上进行了实验,并将其与使用6种不同变异策略的原始DE和4种改进DE进行比较.实验结果表明,提出的算法是一种有效的DE变种,其性能优于其它DE.     

可求解多目标优化问题的演化博弈优化算法

借鉴演化博弈的思想和选择机制,提出了一种新的基于演化博弈的优化算法(EGOA)用于多目标问题的求解.算法框架具备对该类问题的通用性.为了对算法性能进行评估,采用了一组多目标优化问题(MOPs)的测试函数进行实验.实验结果表明,使用本算法搜索得到的演化稳定策略集合能够很好地逼近多目标优化问题的帕累托前沿,与一些经典的演化算法相比具有良好的问题求解能力.     

面向离散优化问题的量子协同演化算法

为解决现有离散优化算法在有限时间内容易出现过早收敛或难以收敛的问题,提出了面向离散优化问题的量子协同演化算法。该算法通过种群初始化策略构建分布均匀的初始种群,并改进粒子群和单点优化算法成为具有不同搜索能力的协同演化策略,进而利用量子旋转门根据种群个体的进化情况自适应地选择合适的演化策略,最后利用精英保持策略避免种群的退化。在标准离散问题和背包问题的测试环境中,各算法的平均收敛精度和实际收敛情况均表明,已提出的算法能够在有限时间内,收敛到精度较高的解,可用于求解具有时效要求的离散优化问题。     

可求解多目标优化问题的演化博弈优化算

借鉴演化博弈的思想和选择机制,提出了一种新的基于演化博弈的优化算法(EGOA)用于多目标问题的求解.算法框架具备对该类问题的通用性.为了对算法性能进行评,采用了一组多目标优化问题(MOPs)测试函数进行实验.实验结果表明,使用本算法搜索得到的演化稳定策略集合能够很好地逼近目标优化问题的帕累托前沿,与一些经典的演化算法相比具有良好的问题求解力.     

DE/EDA: A new evolutionary algorithm for global optimization

Differential evolution (DE) was very successful in solving the global continuous optimization problem. It mainly uses the distance and direction information from the current population to guide its further search. Estimation of distribution algorithm (EDA) samples new solutions from a probability model which characterizes the distribution of promising solutions. This paper proposes a combination of DE and EDA (DE/EDA) for the global continuous optimization problem. DE/EDA combines global information extracted by EDA with differential information obtained by DE to create promising solutions. DE/EDA has been compared with the best version of the DE algorithm and an EDA on several commonly utilized test problems. Experimental results demonstrate that DE/EDA outperforms the DE algorithm and the EDA. The effect of the parameters of DE/EDA to its performance is investigated experimentally.     

Circle detection using discrete differential evolution optimization

This paper introduces a circle detection method based on differential evolution (DE) optimization. Just as circle detection has been lately considered as a fundamental component for many computer vision algorithms, DE has evolved as a successful heuristic method for solving complex optimization problems, still keeping a simple structure and an easy implementation. It has also shown advantageous convergence properties and remarkable robustness. The detection process is considered similar to a combinational optimization problem. The algorithm uses the combination of three edge points as parameters to determine circle candidates in the scene yielding a reduction of the search space. The objective function determines if some circle candidates are actually present in the image. This paper focuses particularly on one DE-based algorithm known as the discrete differential evolution (DDE), which eventually has shown better results than the original DE in particular for solving combinatorial problems. In the DDE, suitable conversion routines are incorporated into the DE, aiming to operate from integer values to real values and then getting integer values back, following the crossover operation. The final algorithm is a fast circle detector that locates circles with sub-pixel accuracy even considering complicated conditions and noisy images. Experimental results on several synthetic and natural images with varying range of complexity validate the efficiency of the proposed technique considering accuracy, speed, and robustness.     

An effective hybrid DE-based algorithm for multi-objective flow shop scheduling with limited buffers

This paper proposes an effective hybrid algorithm based on differential evolution (DE), namely HDE, to solve multi-objective permutation flow shop scheduling problem (MPFSSP) with limited buffers between consecutive machines, which is a typical NP-hard combinatorial optimization problem with strong engineering background. Firstly, to make DE suitable for solving scheduling problems, a largest-order-value (LOV) rule is presented to convert the continuous values of individuals in DE to job permutations. Secondly, after the DE-based exploration, an efficient local search, which is designed based on the landscape of MPFSSP with limited buffers, is applied to emphasize exploitation. Thus, not only does the HDE apply the parallel evolution mechanism of DE to perform effective exploration (global search) in the whole solution space, but it also adopts problem-dependent local search to perform thorough exploitation (local search) in the promising sub-regions. In addition, the concept of Pareto dominance is used to handle the updating of solutions in sense of multi-objective optimization. Moreover, the convergence property of HDE is analyzed by using the theory of finite Markov chain. Finally, simulations and comparisons based on benchmarks demonstrate the effectiveness and efficiency of the proposed HDE.     

Topology optimization of structures using modified binary differential evolution

Differential evolution (DE) is an efficient population based algorithm used to solve real-valued optimization problems. It has the advantage of incorporating relatively simple and efficient mutation and crossover operators. However, the DE operator is based on floating-point representation only, and is difficult to use when solving combinatorial optimization problems. In this paper, a modified binary differential evolution (MBDE) based on a binary bit-string framework with a simple and new binary mutation mechanism is proposed. Two test functions are applied to verify the MBDE framework with the new binary mutation mechanism, and four structural topology optimization problems are used to study the performance of the proposed MBDE algorithm. The experimental studies show that the proposed MBDE algorithm is not only suitable for structural topology optimization, but also has high viability in terms of solving numerical optimization problems.     

A HYBRID METAHEURISTIC FOR THE SINGLE-MACHINE TOTAL WEIGHTED TARDINESS PROBLEM

This article presents a new hybrid algorithm for combinatorial optimization that combines differential evolution (DE) with variable neighborhood search (VNS). DE (a population heuristic for optimization over continuous search spaces) is used as global optimizer for solution evolution guiding the search toward the optimal regions of the search space; VNS (a random local search heuristic based on the systematic change of neighborhood) is used as a local optimizer performing a sequence of local changes on individual DE solutions until a local optimum is found. The effectiveness of a DE-VNS approach is demonstrated on the solution of the single-machine total weighted tardiness scheduling problem. The concepts of Lamarckian and Baldwinian learning are also investigated and discussed. Experiments on known benchmark data sets show that DE-VNS with Lamarckian learning can produce high-quality schedules in a rather short computation time. DE-VNS uses a self-adapted mechanism for tuning the required control parameters, a critical feature rendering it applicable to real-life scheduling problems.     

Remote sensing image subpixel mapping based on adaptive differential evolution

In this paper, a novel subpixel mapping algorithm based on an adaptive differential evolution (DE) algorithm, namely, adaptive-DE subpixel mapping (ADESM), is developed to perform the subpixel mapping task for remote sensing images. Subpixel mapping may provide a fine-resolution map of class labels from coarser spectral unmixing fraction images, with the assumption of spatial dependence. In ADESM, to utilize DE, the subpixel mapping problem is transformed into an optimization problem by maximizing the spatial dependence index. The traditional DE algorithm is an efficient and powerful population-based stochastic global optimizer in continuous optimization problems, but it cannot be applied to the subpixel mapping problem in a discrete search space. In addition, it is not an easy task to properly set control parameters in DE. To avoid these problems, this paper utilizes an adaptive strategy without user-defined parameters, and a reversible-conversion strategy between continuous space and discrete space, to improve the classical DE algorithm. During the process of evolution, they are further improved by enhanced evolution operators, e.g., mutation, crossover, repair, exchange, insertion, and an effective local search to generate new candidate solutions. Experimental results using different types of remote images show that the ADESM algorithm consistently outperforms the previous subpixel mapping algorithms in all the experiments. Based on sensitivity analysis, ADESM, with its self-adaptive control parameter setting, is better than, or at least comparable to, the standard DE algorithm, when considering the accuracy of subpixel mapping, and hence provides an effective new approach to subpixel mapping for remote sensing imagery.     

A self-adaptive differential evolution algorithm for continuous optimization problems

This paper proposes a new self-adaptive differential evolution algorithm (DE) for continuous optimization problems. The proposed self-adaptive differential evolution algorithm extends the concept of the DE/current-to-best/1 mutation strategy to allow the adaptation of the mutation parameters. The control parameters in the mutation operation are gradually self-adapted according to the feedback from the evolutionary search. Moreover, the proposed differential evolution algorithm also consists of a new local search based on the krill herd algorithm. In this study, the proposed algorithm has been evaluated and compared with the traditional DE algorithm and two other adaptive DE algorithms. The experimental results on 21 benchmark problems show that the proposed algorithm is very effective in solving complex optimization problems.     

二次分配问题的改进粒子群优化算法

粒子群优化算法已经成功地应用于求解连续域问题,但是对于离散域问题的求解,尤其涉及组合优化问题的研究和应用还很少。二次分配问题本身是一个离散域问题,因此,使用粒子群算法求解二次分配问题是一个新的研究方向。文章引入交叉策略和变异策略对粒子群优化算法进行改造,使得粒子群优化算法可以用来解决二次分配问题。