ACenterMulti-swarmCooperativeParticleSwarmOptimizationwithRatioandProp资源-CSDN文库

需积分: 9 65 浏览量 2020-09-22 09:57:10 上传评论收藏 335KB PDF 举报

《中心多群协同粒子群优化算法：比率与比例学习》粒子群优化（PSO）由Kennedy和Eberhart在1995年首次提出，作为一种高效的全局优化方法，它与其他元启发式算法和搜索算法（如遗传算法(GA)和差分进化(DE)）一样，在解决一般问题时表现出色。由于其概念简单、效率高，PSO已被广泛应用于各种领域，包括工程优化、机器学习、神经网络训练等。然而，原始的PSO算法在处理复杂优化问题时，可能会遇到收敛速度较快但多样性降低的问题，导致过早收敛，无法充分探索搜索空间的所有可能区域。为了解决这个问题，本文提出了一种中心多群协同粒子群优化算法，结合比率和比例学习（CMCPSO-RP），借鉴了心理学中的认知负荷理论和人类问题解决理论。 CMCPSO-RP算法的核心在于通过中心通信机制收集各个子群的最佳粒子，形成一个潜在解决方案的种群，以防止陷入局部最小值。这种策略提高了全局搜索的能力，但在迭代效率上可能有所牺牲。为了解决这一矛盾，算法引入了比率和比例学习策略，赋予搜索粒子类似人类思考和认知过程的能力。比率设计不仅体现了解决方案质量和速度之间的权衡，还使CMCPSO-RP更符合自然规律中的有序学习原理。在实验部分，CMCPSO-RP算法被应用于多个基准优化问题，与对比算法相比，在所有测试函数上都显示出显著的性能提升。这表明，比率和比例学习策略成功地平衡了算法的探索和利用能力，提高了优化效果。总结来说，CMCPSO-RP算法是一种创新的优化方法，通过中心多群结构和比率比例学习，解决了传统PSO算法的局限性，提高了对复杂优化问题的求解能力。该算法的贡献在于将心理学理论融入到优化算法中，实现了更高效、更具适应性的搜索过程，为未来优化算法的设计提供了新的思路。

资源推荐

资源详情

资源评论

A Center Multi-swarm Cooperative Particle

Swarm Optimization with Ratio

and Proportion Learning

Xuemin Liu, Lili

(&)

, and Jiaoju Ge

(&)

Harbin Institute of Technology (Shenzhen), Shenzhen 518055, China

ximlli@126.com, jiaoge@hit.edu.cn

Abstract. This paper presents a center multi-swarm cooperative PSO with ratio

and proportion learning (CMCPSO-RP), employing two well-known psychol-

ogy theories. In the original MCPSO-CC, the convergence speed can be

accelerated which comes at decreasing the diversity of sub-swarms, suffering

from premature convergence. There is no mechanism to guarantee every pos-

sible region of the search space could be searched. To tackle this problem, all

best particles from each sub-swarm can be collected and sent to master swarm to

maintain a population of potential solutions. This process is less prone to

becoming trapped in local minima, but typically has lower efﬁciency of itera-

tions. To balance the ability of exploration and exploitation, a ratio and pro-

portion learning strategy is proposed by empowering the searching particles

with human-like thinking and cognitive process, inspired by Cognitive Load

Theory and Human Problem Solving Theory. In our approach, a reasonable ratio

design can be not only a way to exhibit a solution quality versus speed tradeoff,

but also make CMCPSO-RP more in line with the laws of regular learning in

nature. Application of the newly developed PSO algorithm on several bench-

mark optimization problems shows a marked improvement in performance over

the comparison algorithms on all test functions.

Keywords: Multi-swarm

 Center communication  Ratio and proportion

learning

1 Introduction

Particle swarm optimization (PSO) was ﬁrst developed by Kennedy and Eberhart in

1995 [1]. For general problems, it provides efﬁcient and satisfactory solutions like other

meta-heuristic methods and search algorithms [2], such as genetic algorithms (GA) [3]

and differential evolution (DE) [4]. Owing to its simple concept and high efﬁciency,

PSO has been used to solve a range of optimization problems [5], including neural

network training [6], data mining [7], feature selection [8], just to name a few. Con-

sidering the fact that the original PSO works better in simple and low-dimensional

search space, various attempts have been made to improve its performance from

aspects of topology, parameter control mechanism, learning strategy and hybridization.

For solving complex multimodal problems, numerous multi-swarm techniques, aiming

to improve the population diversity, are studied, such as dynamic multi-swarm PSO

Y. Tan et al. (Eds.): ICSI 2017, Part I, LNCS 10385, pp. 189–197, 2017.

DOI: 10.1007/978-3-319-61824-1_21

(DMS-PSO) [9], cooperative PSO (CPSO) [10] and multi-swarm cooperative PSO

(MCPSO) [11]. Though these efforts, preventing premature convergence while

retaining the fast-converging feature of PSO is still a challenging task. Hence, it is

necessary to design a method that performs well in controlling the compatibility

between exploration and exploitation. Note that, to our knowledge, few scholars pay

attention to augment PSO from the perspective of cognitive process. Given this,

Human Problem Solving Theory (HPST) [12] and the Cognitive Load Theory

(CLT) [13–15] are incorporated, empowering the searching particles with human -like

thinking. The goals of this work are as follows: (1) Reinterpret learning process with

HPST and CLT in PSO; (2) Design a ratio and proportion learning strategy to balance

convergence rate and diversity; (3) Empirically measure the effect of the learning

method proposed.

The remainder of this article is organized as follows: Intr oduce the framework of

CMCPSO-RP in Sect. 2. Then experiment setting and results are presented in Sect. 3.

Followed by conclusions and future work in Sect. 4.

2 The Proposed MCPSO-RP

2.1 Reinterpret Learning Process with HPST and CLT

HPST explains the nature of problem solving [12]. When handling speciﬁed tasks, a

task analysis is primary, including what processes are used, and what mecha nisms

perform these processes. Then seek an inventory of possible problem solving mech-

anisms. At last, summarize these experiences, select what can be learned from, and the

relation of these to the problem-solving process.

Compared with current learning methods used in PSOs, HPST takes advantage of

logic in learning and ways of thinking. As for the strategy to assign the ratio, we ﬁnd the

answer in CLT, stating that if interactions between many elements must be learn ed, then

intrinsic cognitive load will be high, which will lead to a low efﬁciency of learning. In

contrast, if elements can be learned successively rather than simultaneously, intrinsic

cognitive load will be low and the learning process will be more effective [13].

2.2 CMCPSO-RP

The motivation for developing CMCPSO-RP derives from MCPSO-CC [16]. For the

original MC PSO-CC, while promising, there is still room to improve its performance.

• The center communication mechanism sacriﬁce the diversity.

• It cannot guarantee to search every possible region of the search space.

• The particle updates its velocity with the same opportunity, which is somehow

contrary to the laws of regular learning in natur al ecosystems.

To address above issues, CMCPSO-RP is proposed based on HPST and CLT.

According to CLT and HPS, when facing a new problem, human tend to believe

themselves ﬁrst and try their best to deal with the difﬁculty based on the existing

knowledge, however, if they found it beyond their ability after attempts, they will begin

190 X. Liu et al.

剩余8页未读，继续阅读

评论收藏

内容反馈

高鹰-广州大学

粉丝: 2
资源: 26

A Center Multi-swarm Cooperative Particle Swarm Optimization wit...

最新资源

A Center Multi-swarm Cooperative Particle Swarm Optimization wit...

Convergent Heterogenous Particle Swarm Optimization:Multi-swarm,收敛分析,异构搜索,协同-matlab开发

Particle Swarm Optimization

particle swarm optimisation

particle swarm optimization

Particle-swarm-optimization

Fuzzy-multi-objective particle swarm optimization

Multi-objective particle swarm optimization MOPSO.zip

Multi-Objective Particle Swarm Optimization (MOPSO)多目标的粒子群算法

A novel multi-objective particle swarm optimization with multiple search strategies

Multi-Objective Particle Swarm Optimization ------(MOPSO)

Particle-Swarm-Optimization

A co-evolving framework for robust particle swarm optimization

multi-objective particle swarm optimizers

Particle Swarm Optimization，PSO，粒子群优化算法

A modified particle swarm optimizer

matlab代码粒子群算法-Self-Adaptive-Particle-Swarm-Optimization-for-Large-Scale

Enhanced-Binary-Particle-Swarm-Optimization-(BPSO_bpso

ego-planner-swarm 注释版

mall-swarm是一套微服务商城系统，采用了 Spring Cloud、Spring Boot、

Application of multi-phase particle swarm optimization technique to retrieve the particle size distribution

具有统计变量相互依赖学习的协同粒子群优化器

粒子群协同优化理论与方法

Multi-objective-particle-swarm.zip_多目标_多目标 matlab_多目标matlab_粒子群_

多目标粒子群优化算法MATLAB_matlab 多目标优化pso算法,matlab多目标粒子群

最新资源