Archive for the ‘My Study’ Category
Green function techniques in the treatment of quantum transport at the molecular scale
Wednesday, May 21st, 2008A new review presented by G. Cuniberti.
The theoretical investigation of charge (and spin) transport at nanometer length scales requires the use of advanced and powerful techniques able to deal with the dynamical properties of the relevant physical systems, to explicitly include out-of-equilibrium situations typical for electrical/heat transport as well as to take into account interaction effects in a systematic way. Equilibrium Green function techniques and their extension to non-equilibrium situations via the Keldysh formalism build one of the pillars of current state-of-the-art approaches to quantum transport which have been implemented in both model Hamiltonian formulations and first-principle methodologies. We offer a tutorial overview of the applications of Green functions to deal with some fundamental aspects of charge transport at the nanoscale, mainly focusing on applications to model Hamiltonian formulations.
DWNTs and who’s talking first?
Tuesday, May 6th, 2008Weitao Yang果牛人也
Wednesday, April 9th, 20084月7日的Physcal Review Letters出版了杨伟涛的一篇解决密度泛函理论中由交换关联函数引起的大体系计算的离域和局域误差的文章(PRL,100,146401),我想这篇文章肯定会成为本领域的经典之作,即使不成为经典,也会是一篇非常重要的文章。
事实上,这个问题很多人都注意到了,当然也包括我。只是我没有想到如此简单的方法可以解决。
supercurrent in graphene
Sunday, February 24th, 2008An old paper from Nature predict the supercurrent in graphene:( Nature, 446, 57(2007)):
Graphene has become a model system in condensed matter physics because its charge-carrying particles move at relativistic speeds, in effect behaving as if they are massless. This leads to some peculiar electron transport properties, like the one described in this issue. In a graphene layer sandwiched between two superconducting electrodes, a superconducting current flows at low temperatures. Current is carried either by electrons or by holes, depending on the gate voltage and hence charge density in the graphene layer. Interestingly, a finite supercurrent can flow even when the charge density is zero. These observations shed light on the relativistic phenomenon known as time-reversal symmetry, and on the nature of transport mechanisms in graphene.
Graphene已经成为凝聚态物理学中的一个模型体系,因为其带电荷的粒子能以相对论速度运动,从效果来看,其行为就好像它们没有质量一样。这种现象导致了一些奇特的电子输送性质。在夹在两个超导电极之间的一个Graphene层中,一个超导电流可在低温下流动。电流要么是被电子、要么是被空穴运载的,这取决于门电压,因而也取决于Graphene层中的电荷密度。有趣的是,即使当电荷密度为零时,一个有限的超电流(supercurrent)也可以流动。这些观测结果为研究被称为“时间反转对称”的相对论现象以及Graphene中的电子输送机制提供了线索。
PageRank,马尔科夫链及个人推荐
Sunday, December 9th, 2007Google之所以能如此高效搜索主要是基于其创始人Larry Page和Sergey Brin在Stanford读研究生期间提出的PageRank算法。PageRank完全由WWW的超链接结构所决定,它大约没隔一个月重新计算一次,而与任何网页的实际内容或搜索请求无关。当用户发出搜索请求时,Google找出符合请求的网页,并把他们按PageRank值的大小依次列出。从这个意义上讲,这正是Google公司值得尊敬的地方,做过数值计算的人都知道这么大的矩阵运算实在是非常困难。但是这也是其欺骗了用户的地方,在Google的工具条上有一个查看PageRank值的地方,这个值并不是现在的Rank值。
我们在网络上闲逛的时候,每次都从当前网页随机选择一个超链接进入下一个网页,可以想象这个网页最终将中止于一个没有网页链接出口的网页,或者进入一个由相互链接组成的死循环。这种随机游走在数学上称为马尔科夫链(Markov Chain)或马尔科夫过程(Markov Process)。当这样的随机浏览过程无限进行下去时,某个网页被访问的概率就是它的PageRank值。概率越大,PageRank值就越大。
对于Google公司而言,可以设某个根网页开始,沿一系列超级链接到达的所有网页组成了一个集合,而且这个集合不断的变大。这个集合构成一个矩阵,解这个矩阵特征值就得到了每个网页的PageRank值。比如Sina网的PageRank值是7,俺们学院的PR值是2,本博的PR值为1,本博母校PR值为6等。到目前为止,这个集合已经超过60亿了。这么庞大的计算量对任何人来讲都实在是很夸张。
然而,对于经常使用Google的人来讲,可以发现今年Google搜索的准确度明显不如04年了。我们要找到自己所需要的信息越来越难了,比如刚才我搜索本博的名字得到424项搜索结果,到底那些项是我自己呢?好在跟我同名的人不是很多,搜索我一个朋友恰恰同学得到的信息条数是152000条,很明显要找到我那个朋友的信息还是有点难的。经常光顾本博的Princess Lin同学,在Google上能搜到126000条,要在网上找到她绝对叫大海捞针,我要找到她本来直接到她办公室就好了,当然前提是我知道她的办公室在哪儿。所以我们可以得到这样的结论,信息过载等于没有信息。
要解决这个问题怎么办呢?目前Web 2.0技术似乎为解决这个问题打开了一扇门。但是个性怎么体现呢?比如在化妆品网站上面,网站通常根据销量大小和其他的商业目的给出推荐产品。但是通常每个人的喜好都不一样,怎样调和大众口味和个人喜好,这也是一个需要解决的问题。所以当我们从解决好这个问题之后,给出合适的推荐算法,这绝对具有重要的商业前景。在这方面恰恰同学已经灌了篇水了(Phys.Rev.E, 76, 046115(2007)).
ps. Google的另一个创始人也结婚了,新娘很漂亮!
关于浮点算术
Wednesday, November 14th, 2007今天看了一下blog的访问量,一部分访问者来自美国,这个我还真不知道是谁,很小的访问来自澳大利亚,这自然是xiao ya同学了。就单篇文章而言,以前一篇关于matlab内存管理的文章竟然是访问量最大的,当然多是使用google, baidu上搜索来的,不过这也说明写一些具体的专业方面的技巧还是有意义的。今天开贴聊聊浮点算术。
很多人认为数值分析是关于浮点算术的研究,或者是浮点算术不可预料,更难于理解,其实这种观念是错误的。其实我们在实际计算中只有很少的内容是关于浮点算术的,但在讨论这些问题时,往往浮点算术体现了其强大的工具性和数学上的优美性。
当我们考虑比如加法、乘法等基本算术运算的定义时,便不可避免的涉及实数这样的抽象数学。由于实数的定义包括了极限和无限的概念,所以不适合在实际计算中使用。事实上,Matlab和大多数其它科学计算环境中使用的浮点算术体系一样,包含了一个有限精度的有限数的集合,这会导致舍入、下溢和上溢等现象出现。通常在使用Matlab时,我们不需要考虑这些细节,但偶尔可以根据他们解释浮点数的一些性质和限制。
二十多年前,关于浮点数使用的状况比现在复杂的多,江湖上,各种帮派多如牛毛。那时候没种计算机都有自己的浮点数体系,有的使用二进制,有的使用十进制,甚至还有一种俄罗斯计算机使用三进制算术。而在二进制计算机中,有的以2为基数,有的则使用8和16,并各自采用不同的精度来表示。可能经常玩儿随机数的侠客熟悉这段往事。1985年,IEEE标准委员会和美国国家标准局为二进制浮点数体系共同采纳了ANSI/IEEE 754-1985号标准。这个标准文件是来自大学、计算机制造商、微处理器公司的92位数学家、计算机科学家和工程师组历时十年的工作结晶。至此,二进制浮点数体系终于一统江湖。
1985年以后的计算机都采用IEEE标准的浮点算术体系,这并不意味着他们有同样的计算结果,因为这个标准内还允许有少量的变化。这跟武当太极剑一样,每个剑客使用的招式可能都稍有差别,但是,太极剑剑意还是一样的。的确,我们现在有一个与计算机类型无关的模型,可以研究浮点算术的行为。
具体到Matlab使用的浮点数体系,有其特殊性。
不过我的咖啡喝完了,故事下次接着讲。
逾渗电流
Tuesday, October 30th, 2007在以前的一篇文章中提到,arXiv:0706.2968利用随机电阻器网络来分析电导对掺杂和无序的标度依赖,以及磁阻和移相率。发表在这一期Phys.Rev.Lett.上了( Phys.Rev.Lett., 99, 176801).
A mathematical and computational review of Hartree-Fock SCF methods in Quantum Chemistry
Friday, October 19th, 2007Pablo Echenique, J. L. Alonso give a review of the fundamental topics of Hartree-Fock theory in Quantum Chemistry. From the molecular Hamiltonian, using and discussing the Born-Oppenheimer approximation, we arrive to the Hartree and Hartree-Fock equations for the electronic problem. Special emphasis is placed in the most relevant mathematical aspects of the theoretical derivation of the final equations, as well as in the results regarding the existence and uniqueness of their solutions. All Hartree-Fock versions with different spin restrictions are systematically extracted from the general case, thus providing a unifying framework. Then, the discretization of the one-electron orbitals space is reviewed and the Roothaan-Hall formalism introduced. This leads to a exposition of the basic underlying concepts related to the construction and selection of Gaussian basis sets, focusing in algorithmic efficiency issues. Finally, we close the review with a section in which the most relevant modern developments (specially those related to the design of linear-scaling methods) are commented and linked to the issues discussed. The whole work is intentionally introductory and rather self-contained, so that it may be useful for non experts that aim to use quantum chemical methods in interdisciplinary applications. Moreover, much material that is found scattered in the literature has been put together here to facilitate comprehension and to serve as a handy reference.
It will appeared in Molecular Physics.
Statistical physics of social dynamics
Friday, October 19th, 2007A New Review about social dynamics:
Statistical physics has proven to be a very fruitful framework to describe phenomena outside the realm of traditional physics. The last years have witnessed the attempt by physicists to study collective phenomena emerging from the interactions of individuals as elementary units in social structures. Here we review the state of the art by focusing on three major research lines i.e., opinion, cultural and language dynamics. In addition we discuss other social phenomena, such as crowd behavior, hierarchy formation, human dynamics, social spreading. We highlight the connections between these problems and other, more traditional, topics of statistical physics. We also emphasize the comparison of model results with empirical data from social systems.
It is submitted to Rev.Mod.Phys. .

