基于信息再利用的灰色系统GM(1.1)模型建模方法及应用

目的:寻找新的灰色系统GM(1.1)模型建模方法,建立拟合精度与预测精度较高的GM(1.1)模型.方法:在邓聚龙教授建模方法的基础上,用基于信息再利用的方法,建立新的灰色系统GM(1.1)模型.结果:用基于信息再利用的灰色系统GM(1.1)模型建模方法建立的GM(1.1)模型,其拟合精度与预测精度不但优于传统方法建立的GM(1.1)模型,而且优于其他改进方法建立的GM(1.1)模型.结论:基于信息再利用的灰色系统GM(1.1)模型建模方法不但建模过程简单适用,而且其建立的GM(1.1)模型拟合精度与预测精度优于其他改进方法建立的GM(1.1)模型,因而具有广泛的应用价值.     

基于多源数据特征驱动和多尺度分析的PM2.5混合预测研究

精准把握PM_2.5污染的动态演变规律对政府和企业的大气污染防治决策至关重要.因此,文章提出了基于多源数据特征驱动及多尺度分析的混合预测建模框架,以提高PM_2.5预测精度.预测建模框架分为:1)多源数据分析,有效融合与PM_2.5污染相关的气象、污染、舆情等多源数据;2)多尺度分析,通过多元经验模态分解技术(MEMD)将多源数据分解成不同模态下的预测特征;3)混合预测分析,有序结合计量和机器学习模型,集成各模态预测值为最终结果.文章以北京市PM_2.5为研究案例结果表明:1)文章提出的混合模型的预测精度优于所有的基准模型;2)微博个数和情感能够叠加提升PM_2.5预测精度,且优于单因素预测结果;3)引入MEMD分解的模型精度显著高于基准模型.     

基于离散指数函数优化GM(1,1)模型的再优化

分析了基于离散指数函数优化的GM(1,1)模型虽然大幅度提高建模的精度,但在构造新背景值过程中仍存在误差的原因,并针对此原因提出了进一步优化此背景值的方法,从而再次提高了建模的精度.经过严格理论验证该模型具有白化指数重合性,所以既适合用于低增长指数序列建模,也适合用于高增长指数序列建模.同时通过大量的数据模拟,并与原GM(1,1)模型及其基于离散指数函数优化的模型对比,发现本文优化的GM(1,1)新模型有非常高的模拟精度和预测精度.     

基于灰色系统的逻辑斯蒂模型的建模方法

给出了基于灰色系统的逻辑斯蒂(logisitic)模型的建模方法.用实例验证了基于灰色系统的逻辑斯蒂(logisitic)模型的建模过程简便适用,拟合精度较高.     

晶格反演的嵌入原子法模型及其应用

讨论一种基于晶格反演的嵌入原子法模型 ,其特点是未知函数的参数化是对原子间对势和单原子电荷密度函数的晶格和 ,而不是对原子间对势和单个原子的电荷密度函数进行的 ,其最大优点在于待定参数可以用物理输入解析地表达出来 ,而不是像大多数埋入原子法模型中那样通过拟合来确定 .给出了它在计算点缺陷 (自间隙杂质和空位原子 )和晶体表面形成能方面的应用 .     

GM(1,1)模型预测精度仿真分析

基于GM(1,1)模型的灰色预测具有独特优点.从GM(1,1)建模数据的选择入手,应用数值仿真方法,针对具有不同发展系数和偏离度的大量模拟序列研究了建模维数与预测精度之间的关系.研究结果给出了不同情况下的最佳建模维数和预测精度期望值,为GM(1,1)建模提供了有益的指导信息.     

过渡金属Ni晶界的电子结构

本文基于重位点阵模型,用分子动力学方法弛豫界面原子,给出倾转∑5晶界原子坐标.在此基础上,用递推方法计算了过渡金属Ni的晶态及含晶界态的电子结构,给出了晶界能、键序及原子间相互作用,以及电荷转移信息.同时,定义了界面电子活性.结果表明:预弛豫的∑5晶界是一类可能出现的结构,且有俘获电子的作用.此外,计算还表明,与晶态相比,界面原子间力有减弱的倾向,这意味着界面原子属弱连接结构.     

优化的非等距GM(1,1)模型在基坑周边地表沉降预测中的应用

由于影响基坑沉降的因素较多,并且在实际工作中监测数据存在非等距的情况,通过传统的非等距GM(1,1)模型的建模原理分析了其预测精度低的原因,同时指出背景值是影响非等距GM(1,1)模型精度的关键因素之一.在此基础上,提出运用Newton插值法和Newton-Cotes求积公式优化背景值,结合工程实例,表明优化后的非等距GM(1,1)模型在沉降预测中的有效性.     

一类准指数序列的模型优化

在传统GM(1,1)模型基础上,结合最小二乘法原理提出:对本身已具有准指数规律的原始序列直接进行建模,并在此基础上对新模型背景值进行适当优化.克服传统GM(1,1)模型建模过程中的盲目性,并提高了拟合与预测精度.     

基于上凸序列的GM(1,1)模型及其优化

针对GM(1,1)模型对上凸序列建模时会出现误差较大的情况进行了研究.首先分析了GM(1,1)对上凸序列建模时的残差变化规律,然后通过分析得出了残差变化规律的精确描述,同时证明了残差序列的几个性质定理.基于残差序列的性质定理提出了基于上凸序列建模的残差修正GM(1,1)模型.将新模型与多种改进的GM(1,1)模型进行对比,实证结果表明新模型具有很高的模拟预测精度,并且适用于一切上凸序列的建模.     

Properties of learning models in collective action: Rationality of backward-looking players

This article intends to clarify properties of learning models in simulation studies and to conduct a comparison of preceding learning models. Learning models are often used in many simulation studies, but there is no uniform rule of learning. We introduce three technical properties (monotonicity, condition of probability, neutrality) and three rational properties (rationality is fixed situations, rationality in first order stochastic domination, rationality with risk preference in stocahstic situations). We examine Michael Macy's model, the Erev & Roth model, and some others. We find that these models have different properties. Though learning is treated as one of the solutions of social dilemma from the results of Macy's model (Kollock, 1998), Macy's model is peculiar learning model. Learning is not always a solution of social dilemma. A comparison of learning models from a uniform point of view clarifies the properties of each model, and helps to probe conformity of a learning model and human behavior.     

Long-lived discrete breathers in free-standing graphene

Intrinsic localized modes or discrete breathers (DBs) are investigated by molecular dynamics simulations in free-standing graphene. DBs are generated either through thermal quenching of the graphene lattice or by proper initialization, with frequencies and lifetimes sensitively depending on the interatomic potential describing the carbon-carbon interaction. In the most realistic scenario, for which temperature-dependent molecular dynamics simulations in three dimensions using a graphene-specific interatomic potential are performed, the DBs lifetimes increase to hundreds of picoseconds even at relatively high temperatures. These lifetimes are much higher than those anticipated from earlier calculations, and may enable direct breather observation in Raman spectroscopy experiments. Our simulations provide clear estimation for the temperatures in which DBs are expected to be thermally excited (1500–2000 K) representing a step forward for understanding the nonlinear physics of graphene and designing experiments in order to detect DBs, with possible impact in graphene-based future technological applications.     

Can Monte-Carlo Tree Search learn to sacrifice?

One of the most basic activities performed by an intelligent agent is deciding what to do next. The decision is usually about selecting the move with the highest expectation, or exploring new scenarios. Monte-Carlo Tree Search (MCTS), which was developed as a game playing agent, deals with this exploration–exploitation ‘dilemma’ using a multi-armed bandits strategy. The success of MCTS in a wide range of problems, such as combinatorial optimisation, reinforcement learning, and games, is due to its ability to rapidly evaluate problem states without requiring domain-specific knowledge. However, it has been acknowledged that the trade-off between exploration and exploitation is crucial for the performance of the algorithm, and affects the efficiency of the agent in learning deceptive states. One type of deception is states that give immediate rewards, but lead to a suboptimal solution in the long run. These states are known as trap states, and have been thoroughly investigated in previous research. In this work, we study the opposite of trap states, known as sacrifice states, which are deceptive moves that result in a local loss but are globally optimal, and investigate the efficiency of MCTS enhancements in identifying this type of moves.     

演化博弈与自组织合作

主要研究复杂网络上的演化博弈.首先研究具有社团结构的无标度网络上的演化囚徒困境博弈及Newman-Watts小世界网络中异质性对合作演化的影响.然后考察了在不同合作者和作弊者初始分布配置情况下,不同的初始比例条件对合作水平的影响,且在社会网络上研究了雪堆博弈中的合作演化.进一步地,讨论了网络拓扑和博弈动力学的共同演化问题和网络上演化囚徒困境中的强化学习问题.最后给出了复杂网络上演化博弈论的未来发展方向与应用前景.     

A multiscale molecular switch model

A first-principles mathematical simulation of molecular switching of the isomerization reaction in a naphthalocyanine molecule on the basis of a current-induced tunnel microscope is presented. The Gibbs free energy surface and the reaction pathway are analyzed in terms of metadynamics using the numerical quantum mechanical Car-Parrinello molecular dynamics (CPMD) code. The calculations were performed on an IBM Blue Gene/P supercomputer at Moscow State University’s Faculty of Computational Mathematics and Cybernetics. The height of the energy barrier that must be overcome to achieve the isomerization reaction is found along an introduced coordination variable. A multiscale model of Ehrenfest molecular dynamics states is proposed on the basis of an excited electron. The model uses the coordination direction of the reaction on the free energy surface obtained in calculations for a complete molecule.     

Special Issue Article: A Look at Technology's Role in Professional Development of Mathematics Teachers at the Middle School Level

In last month's issue of School Science and Mathematics, Glenda Lappan addressed the dilemma of supporting teachers in continuing to grow professionally in four domains: Learning more mathematics content; improving their pedagogical skills; improving assessment skills; and addapting the curriculum to the needs of their students. Though it is clear that this kind of learning is best done in the context of what is going on in the classroom, teachers' schedules and other problems make it difficult to implement a coherent classroom based, professional development program. In this article, two more knowledge domains are added to this mix: the ability to (a) use technology and (b) teach with technology effectively. Since the process of learning and teaching is a dynamic one, the author describes his vision of the classroom as a laboratory where teachers get to practice and improve in these six areas and get feedback from an audience of their peers. Reflections are based on a current project in Paterson, New Jersey where the author helps middle school teachers use computer software to improve their mathematics teaching and learning.     

Lower Bound for the Interatomic Distance in Lennard-Jones Clusters

We prove in this article lower bounds and upper bounds for the interatomic distance in cluster of atoms minimizing the Lennard-Jones energy. Our main result is in dimension three, but we also prove it in the two-dimensional case, since it seems interesting from a theoretical point of view.     

Derivation of macroscopic relations of the elasticity of complex crystal lattices taking into account the moment interactions at the microlevel

A discrete mechanical model of a complex crystal lattice is proposed which contains particles possessing both translational and rotational degrees of freedom and which interact with one another by means of forces and moments. The transition to a continuum model of a crystal lattice is performed using the long-wave approximation, and, at the same time, it is shown that the dynamics of the continual model are described by the equations of the macroscopic moment theory of elasticity. Expressions are obtained for the macroscopic stiffness tensors which depend on the stiffness tensors of the interatomic bonds and the vectors determining the lattice geometry. A transition to the moment less theory of elasticity is made and it is shown that the macroscopic moduli of elasticity of the moment less theory depend both on the forces and the torque characteristics of the interatomic interaction. The stiffnesses of the interatomic bonds in a layer of graphite are calculated and it is shown that the transverse stiffness of an interatomic bond is comparable with the longitudinal stiffness, that is, a covalent bond is substantially non- central, which is only possible when there are torque interactions at the microlevel.     

Balancing mathematics education research and the NCTM standards

The release of the Principles and Standards for School Mathematics in the United States by the National Council of Teachers of Mathematics (NCTM) brought to the forefront the debate of whether research should determine the validity of the espoused Standards? Or conversely whether the Standards should influence the research agenda of the mathematics education community? How should university teacher educators address this issue? Should pre-service and practicing teachers blindly, accept the Standards as well as the research, or do we cultivate the critical thinking skills that will allow preparing teachers to resolve this dilemma? In this article a university mathematics educator and an idealistic pre-service elementary teacher try to resolve the dilemma of balancing the Standards with research and personal beliefs about the teaching and learning of mathematics.