无环境信息下多尺度网格细胞群空间表征模型

针对无环境信息条件下网格细胞空间表征问题,提出一种多尺度网格细胞群的空间表征模型.通过减少吸引子网络模型网格细胞的偏好朝向,使网格细胞具有特定的感应方向,仅对特定方向上的速度产生响应.基于改进的吸引子网络模型,构建多尺度网格细胞群,并对空间进行表征.仿真结果表明,改进后的网格细胞模型只响应相应方向的速度,准确整合速度信...     

一种新的进化策略及其全局收敛性

在求解单峰最优化问题算法的基础上,给出了一种新的进化策略.针对连续函数优化问题,利用中心极限定理,在较弱的条件下,首先证明了基于均匀分布的(μ λ)-ES算法依概率收敛,然后给出了采用一般连续性随机变量作为变异算子的(μ λ)-ES算法依概率收敛的证明.数值结果表明:采用基于均匀分布的进化策略求解维数较高的连续函数优化问题能够快速有效地收敛到全局最优解.     

创业投资辛迪加网络中控制权的演化博弈分析

创业投资的高风险性和信息不对称性使创业投资中控制权分配问题逐渐成为关系到风险投资项目成败的关键。本文运用演化博弈方法,分析控制权在辛迪加网络中两类创业投资家之间分配的博弈过程,得出在不同控制权分配比例下相应的演化稳定策略。在此结果上表明即期博弈中,创业投资家的私人收益和项目产出分配值的对比关系与控制权的分配密切相关。     

The saws of sawflies: exploring the morphology of the ovipositor in Tenthredinoidea (Insecta: Hymenoptera), with emphasis on Nematinae

The variation in ovipositor sculpture and additional features of the ovipositor apparatus in Tenthredinoidea is explored. Two outgroups and 85 species of tenthredinoids were examined, the emphasis of the taxon sampling being on Nematinae. Twenty-four morphological characters where scored and the data set was analysed under different weighting conditions. The topologies derived from analysing the character set in TNT are poorly resolved and highly unstable when analytical settings are changed. Only a few clades, the Tenthredinoidea s. str., the Diprionidae, the Cimbicidae and the Argidae + Pergidae are consistently retrieved. The functional implications of the variation in ovipositor morphology within Tenthredinidae is discussed in relation to the qualities of the host plant, e.g. the hardness and the depth of penetration in plant tissue.     

Predatory behaviour in Deltochilum: convergent evolution or a primitive character within a clade?

We present data on predatory behaviour in Deltochilum cupreicolle viridescens, which belongs to the subgenus Aganhyboma, and we contrast it with the behaviour of a species from the subgenus Deltohyboma. By this comparison we can investigate whether the predatory behaviour in Deltochilum could have appeared independently in these two subgenera, or if the behaviour is so similar that it suggests a single origin in a possible ancestor exclusive to those two groups. These laboratory studies on the behaviour of Deltochilum c. viridescens individuals on diplopods are described and divided into the following stages: (1) approach and recognition of the diplopod; (2) attack and death of the prey; (3) on-site feeding; (4) transport of the diplopod. The results found here strengthen the hypothesis of a primitive and highly specialized behaviour within a possible clade formed by the predatory species of Diplopoda.     

基于配置约束处理的产品配置设计方法

为了解决复杂的配置设计问题，提出了一种集成配置设计方法．该方法包括3个相互协作的过程：通过实例匹配技术从现有设计实例库中搜索满意设计解；通过领域知识的支持解决具有强烈知识依赖性的模块配置问题及关键参数的赋初值问题；通过进化算法解决配置设计过程中参数的求解和优化问题．针对参数求解和优化过程中的一些复杂配置约束，在传统进化算法的基础上提出了约束校核操作和再设计规则的配置约束处理方法．最后以蜗杆减速器为实例，验证了所提出的基于配置约束处理的产品配置设计方法能够快速给出良好的设计解，从而为复杂产品的配置设计问题提供求解方案．     

自主机器人人工进化方法研究

在介绍人工进化研究状况的基础上 ,着重阐述了进化机器人学的关键问题 :基本概念、进化技术、进化计算和实验研究方法等 ,探索了进化机器人学未来的研究趋势及所面临的挑战 .     

Dynamic Niching Genetic Algorithm with Data Attraction for Automatic Clustering

A genetic clustering algorithm was developed based on dynamic niching with data attraction. The algorithm uses the concept of Coulomb attraction to model the attraction between data points. Then, the niches with data attraction are dynamically identified in each generation to automatically evolve the optimal number of clusters as well as the cluster centers of the data set without using cluster validity functions or a variance-covariance matrix. Therefore, this clustering scheme does not need to pre-specify the number of clusters as in existing methods. Several data sets with widely varying characteristics are used to demonstrate the superiority of this algorithm. Experimental results show that the performance of this clustering algorithm is high, effective, and flexible.     

从思维方式探讨离散数学的结构内涵

结合离散数学的具体内容,选取其中几个较难深入理解的问题,从思维方式的角度探讨其结构内涵,并寻究其所蕴含的方法本源,以期在思维层面上能有更深广的开拓,并且由此指导教学.     

基于遗传算法的最小误工任务数问题求解

为了解决最小误工任务数问题（MTTP），将遗传算法引入该问题的求解中，基于惩罚函数。讨论了MTTP的遗传算法求解方法。并对genGA，ssGA，cGA三种演化式算法求解MTTP的实验运行结果进行分析比较，得出在解决大规模的MTTP时，genGA明显优于另两种演化式算法。