组合生成算法与多元线性回归相结合用于近红外光谱波长的优选

分立波长型近红外光谱分析仪是光谱分析仪器中较为普及的一种快速成份定量分析仪。如滤光片型、发光二极管型等。该类分析仪器研发的一个主要问题是如何针对于待测物质主要成份进行近红外光谱解析。找到最优定标波长组合用于建立稳健的定标模型。常用的波长选择方法为相关光谱结合逐步多元线性回归方法,该方法依据各参与定标波长所对应的t检验值进行最优定标波长的判别,但在实际应用中定标模型的定标精度和预测精度相差较大,具有很大的不准确性。为了实现定标波长的优选引入了组合数学中的组合生成算法。可以在较短的时间内完成最优波长组合的选取,结果是令人满意的。     

模糊多属性群决策一致性分析研究

群决策过程中各成员可能以不同的方式给出决策信息,讨论了四种不同决策信息的统一方法,设置了模糊环境下进行一致性判断的准则,介绍了梯形模糊数在群决策过程中的运算方法,举例说明了这种方法的应用步骤.     

基于累积前景理论和VIKOR的毕达哥拉斯犹豫模糊风险型多属性决策方法

针对属性权重未知,且属性值为毕达哥拉斯犹豫模糊数(PHFN)的风险型多属性决策问题,考虑到决策者的有限理性行为,提出基于累积前景理论(CPT)和多准则妥协优化解(VIKOR)的决策方法。首先,定义PHFN的分散率,并构建优化模型确定属性权重。其次,将CPT融入PHFN环境,定义PHFN的价值函数,并结合决策权重函数计算方案在各属性下的综合前景值。进一步,构建综合前景值矩阵,在此基础上运用VIKOR法确定方案排序。最后,通过风险投资项目选择的应用案例说明所提方法是可行、有效的。     

一种新型风险型多指标决策方法研究

传统的风险型多指标决策模型没有考虑决策者对风险的态度,而决策者对风险的态度会影响决策的结果,针对这一问题文章在累积前景理论与灰色关联方法的基础上,提出一种考虑决策者风险偏好的风险型多指标决策的方法.该方法首先利用极差化法对风险决策矩阵进行规范化处理,并在此基础上构造出最优与最劣方案;然后利用累积前景理论与灰色关联方法构建前景值函数,并给出利用灰色关联思想确定指标权重的方法与步骤;最终求出各个方案的综合前景值并进行排序选优.通过某电信运营商对管道资源建设方案选择的实例分析说明了方法的可行性与有效性.     

Single Sample Path-Based Optimization of Markov Chains

Motivated by the needs of on-line optimization of real-world engineering systems, we studied single sample path-based algorithms for Markov decision problems (MDP). The sample path used in the algorithms can be obtained by observing the operation of a real system. We give a simple example to explain the advantages of the sample path-based approach over the traditional computation-based approach: matrix inversion is not required; some transition probabilities do not have to be known; it may save storage space; and it gives the flexibility of iterating the actions for a subset of the state space in each iteration. The effect of the estimation errors and the convergence property of the sample path-based approach are studied. Finally, we propose a fast algorithm, which updates the policy whenever the system reaches a particular set of states and prove that the algorithm converges to the true optimal policy with probability one under some conditions. The sample path-based approach may have important applications to the design and management of engineering systems, such as high speed communication networks.This work was supported in part by     

Modelling decision making and learning by low latitude hunter gatherers

Recent anthropological studies have demonstrated that low latitude ‘encounter’ foragers exploit their environments in energetically very efficient manners and closely track the environment as it changes. The paper begins to investigate how they manage to do this by proposing a simple decision making and learning rule developed from an evolutionary ecological basis. Having described the mathematical model the paper refers to simulation studies exploring this model which suggest that some of the seemingly complex aspects of hunter gatherer behaviour may result from the use of simple decision making and learning processes.     

Interaction of trip decisions and traffic systems dynamics

A major difficulty in the study of large-scale complex systems involving human decision-making with non-linear collective effects is that of obtaining pertinent data at the desired level of detail. A controlled experiment involving real commuters in a hypothetical computer simulated traffic system is described as an alternative approach of conducting observational studies to support the modelling of such complex dynamic interactive decision systems. An overview of the principal characteristics of the traffic system's evolution obtained in this experiment is presented. Also illustrated are the features of a proposed behavioral framework where users are viewed as boundedly rational seekers of a satisfactory choice outcome.     

层次分析法权重计算方法分析及其应用研究

介绍层次分析法的基本概念,同时也分析了层次分析法权重的计算方法及应用,层次分析法的计算方法有四种方法:几何平均法、算术平均法、特征向量法、最小二乘法,以往的文献利用层次分析法解决实际问题时,都是采用其中的某一种方法求权重向量,不同的方法会导致结果有些偏差,将对一具体实例,采用四种计算方法分别得出权重向量再进行排序和综合分析,这样可以避免采用单一方法所产生的偏差,得出的结论将更全面、更有效。     

需求弹性在价格决策中的应用

将着眼点由总收益改变为总利润,对于需求弹性用于价格决策,利用微积分方法,分5种情况进行推导,得到的结论修正和补充了现行"经济数学"、"市场营销学"和"管理经济学"等教科书里讲的内容,并在此基础上进一步研究了最佳调价量问题.     

An Edge Server Placement Method Based on Reinforcement Learning

In mobile edge computing systems, the edge server placement problem is mainly tackled as a multi-objective optimization problem and solved with mixed integer programming, heuristic or meta-heuristic algorithms, etc. These methods, however, have profound defect implications such as poor scalability, local optimal solutions, and parameter tuning difficulties. To overcome these defects, we propose a novel edge server placement algorithm based on deep q-network and reinforcement learning, dubbed DQN-ESPA, which can achieve optimal placements without relying on previous placement experience. In DQN-ESPA, the edge server placement problem is modeled as a Markov decision process, which is formalized with the state space, action space and reward function, and it is subsequently solved using a reinforcement learning algorithm. Experimental results using real datasets from Shanghai Telecom show that DQN-ESPA outperforms state-of-the-art algorithms such as simulated annealing placement algorithm (SAPA), Top-K placement algorithm (TKPA), K-Means placement algorithm (KMPA), and random placement algorithm (RPA). In particular, with a comprehensive consideration of access delay and workload balance, DQN-ESPA achieves up to 13.40% and 15.54% better placement performance for 100 and 300 edge servers respectively.