S center dot center dot center dot O chalcogen bonding in sulfa drugs: insights from multipole charge density and X-ray wavefunction of acetazolamide

Experimental charge density analysis combined with the quantum crystallographic technique of X-ray wavefunction refinement (XWR) provides quantitative insights into the intra-and intermolecular interactions formed by acetazolamide, a diuretic drug. Firstly, the analysis of charge density topology at the intermolecular level shows the presence of exceptionally strong interaction motifs such as a DDAA-AADD (D-donor, A-acceptor) type quadruple hydrogen bond motif and a sulfonamide dimer synthon. The nature and strength of intra-molecular S center dot center dot center dot O chalcogen bonding have been characterized using descriptors from the multipole model (MM) and XWR. Although pure geometrical criteria suggest the possibility of two intra-molecular S center dot center dot center dot O chalcogen bonded ring motifs, only one of them satisfies the ``orbital geometry'' so as to exhibit an interaction in terms of an electron density bond path and a bond critical point. The presence of `s-holes' on the sulfur atom leading to the S center dot center dot center dot O chalcogen bond has been visualized on the electrostatic potential surface and Laplacian isosurfaces close to the `reactive surface'. The electron localizability indicator (ELI) and Roby bond orders derived from the `experimental wave function' provide insights into the nature of S center dot center dot center dot O chalcogen bonding.

l(1)-K-SVD: A robust dictionary learning algorithm with simultaneous update

We develop a new dictionary learning algorithm called the l(1)-K-svp, by minimizing the l(1) distortion on the data term. The proposed formulation corresponds to maximum a posteriori estimation assuming a Laplacian prior on the coefficient matrix and additive noise, and is, in general, robust to non-Gaussian noise. The l(1) distortion is minimized by employing the iteratively reweighted least-squares algorithm. The dictionary atoms and the corresponding sparse coefficients are simultaneously estimated in the dictionary update step. Experimental results show that l(1)-K-SVD results in noise-robustness, faster convergence, and higher atom recovery rate than the method of optimal directions, K-SVD, and the robust dictionary learning algorithm (RDL), in Gaussian as well as non-Gaussian noise. For a fixed value of sparsity, number of dictionary atoms, and data dimension, l(1)-K-SVD outperforms K-SVD and RDL on small training sets. We also consider the generalized l(p), 0 < p < 1, data metric to tackle heavy-tailed/impulsive noise. In an image denoising application, l(1)-K-SVD was found to result in higher peak signal-to-noise ratio (PSNR) over K-SVD for Laplacian noise. The structural similarity index increases by 0.1 for low input PSNR, which is significant and demonstrates the efficacy of the proposed method. (C) 2015 Elsevier B.V. All rights reserved.

Quantum stochastic calculus associated with quadratic quantum noises

We first study a class of fundamental quantum stochastic processes induced by the generators of a six dimensional non-solvable Lie dagger-algebra consisting of all linear combinations of the generalized Gross Laplacian and its adjoint, annihilation operator, creation operator, conservation, and time, and then we study the quantum stochastic integrals associated with the class of fundamental quantum stochastic processes, and the quantum Ito formula is revisited. The existence and uniqueness of solution of a quantum stochastic differential equation is proved. The unitarity conditions of solutions of quantum stochastic differential equations associated with the fundamental processes are examined. The quantum stochastic calculus extends the Hudson-Parthasarathy quantum stochastic calculus. (C) 2016 AIP Publishing LLC.

Mapping Closure Approximation for Reactive Scalars in Random Flows

The Mapping Closure Approximation (MCA) approach is developed to describe the statistics of both conserved and reactive scalars in random flows. The statistics include Probability Density Function (PDF), Conditional Dissipation Rate (CDR) and Conditional Laplacian (CL). The statistical quantities are calculated using the MCA and compared with the results of the Direct Numerical Simulation (DNS). The results obtained from the MCA are in agreement with those from the DNS. It is shown that the MCA approach can predict the statistics of reactive scalars in random flows.

Logarithmic Potential Theory on Riemann Surfaces

We develop a logarithmic potential theory on Riemann surfaces which generalizes logarithmic potential theory on the complex plane. We show the existence of an equilibrium measure and examine its structure. This leads to a formula for the structure of the equilibrium measure which is new even in the plane. We then use our results to study quadrature domains, Laplacian growth, and Coulomb gas ensembles on Riemann surfaces. We prove that the complement of the support of the equilibrium measure satisfies a quadrature identity. Furthermore, our setup allows us to naturally realize weak solutions of Laplacian growth (for a general time-dependent source) as an evolution of the support of equilibrium measures. When applied to the Riemann sphere this approach unifies the known methods for generating interior and exterior Laplacian growth. We later narrow our focus to a special class of quadrature domains which we call Algebraic Quadrature Domains. We show that many of the properties of quadrature domains generalize to this setting. In particular, the boundary of an Algebraic Quadrature Domain is the inverse image of a planar algebraic curve under a meromorphic function. This makes the study of the topology of Algebraic Quadrature Domains an interesting problem. We briefly investigate this problem and then narrow our focus to the study of the topology of classical quadrature domains. We extend the results of Lee and Makarov and prove (for n ≥ 3) c ≤ 5n-5, where c and n denote the connectivity and degree of a (classical) quadrature domain. At the same time we obtain a new upper bound on the number of isolated points of the algebraic curve corresponding to the boundary and thus a new upper bound on the number of special points. In the final chapter we study Coulomb gas ensembles on Riemann surfaces.

Elimination of zero-order diffraction in digital holography

A simple method to suppress the zero-order diffraction in the reconstructed image of digital holography is presented. In this method, the Laplacian of a detected hologram is used instead of the hologram itself for numerical reconstruction by computing the discrete Fresnel integral. This method can significantly improve the image quality and give better resolution and higher accuracy of the reconstructed image. The main advantages of this method are its simplicity in experimental requirements and convenience in data processing. (C) 2002 Society of Photo-optical Instrumentation Engineers.

Métodos espectrais de agrupamento

Os métodos espectrais são ferramentas úteis na análise de dados, sendo capazes de fornecer informações sobre a estrutura organizacional de dados. O agrupamento de dados utilizando métodos espectrais é comumente baseado em relações de similaridade definida entre os dados. O objetivo deste trabalho é estudar a capacidade de agrupamento de métodos espectrais e seu comportamento, em casos limites. Considera-se um conjunto de pontos no plano e usa-se a similaridade entre os nós como sendo o inverso da distância Euclidiana. Analisa-se a qual distância mínima, entre dois pontos centrais, o agrupamento espectral é capaz de reagrupar os dados em dois grupos distintos. Acessoriamente, estuda-se a capacidade de reagrupamento caso a dispersão entre os dados seja aumentada. Inicialmente foram realizados experimentos considerando uma distância fixa entre dois pontos, a partir dos quais os dados são gerados e, então, reduziu-se a distância entre estes pontos até que o método se tornasse incapaz de efetuar a separação dos pontos em dois grupos distintos. Em seguida, retomada a distância inicial, os dados foram gerados a partir da adição de uma perturbação normal, com variância crescente, e observou-se até que valor de variância o método fez a separação dos dados em dois grupos distintos de forma correta. A partir de um conjunto de pontos obtidos com a execução do algoritmo de evolução diferencial, para resolver um problema multimodal, testa-se a capacidade do método em separar os indivíduos em grupos diferentes.

Decentralised minimal-time consensus

This study considers the discrete-time dynamics of a network of agents that exchange information according to the nearest-neighbour protocol under which all agents are guaranteed to reach consensus asymptotically. We present a fully decentralised algorithm that allows any agent to compute the consensus value of the whole network in finite time using only the minimal number of successive values of its own history. We show that this minimal number of steps is related to a Jordan block decomposition of the network dynamics and present an algorithm to obtain the minimal number of steps in question by checking a rank condition on a Hankel matrix of the local observations. Furthermore, we prove that the minimal number of steps is related to other algebraic and graph theoretical notions that can be directly computed from the Laplacian matrix of the graph and from the underlying graph topology. © 2011 IEEE.

Decentralised minimum-time consensus

We consider the discrete-time dynamics of a network of agents that exchange information according to a nearest-neighbour protocol under which all agents are guaranteed to reach consensus asymptotically. We present a fully decentralised algorithm that allows any agent to compute the final consensus value of the whole network in finite time using the minimum number of successive values of its own state history. We show that the minimum number of steps is related to a Jordan block decomposition of the network dynamics, and present an algorithm to compute the final consensus value in the minimum number of steps by checking a rank condition of a Hankel matrix of local observations. Furthermore, we prove that the minimum number of steps is related to graph theoretical notions that can be directly computed from the Laplacian matrix of the graph and from the minimum external equitable partition. © 2013 Elsevier Ltd. All rights reserved.

Synchronization with partial state feedback on SO(n)

In this paper we consider the problem of constructing a distributed feedback law to achieve synchronization for a group of k agents whose states evolve on SO(n) and which exchange only partial state information along communication links. The partial state information is given by the action of the state on reference vectors in ℝn. We propose a gradient based control law which achieves exponential local convergence to a synchronization configuration under a rank condition on a generalized Laplacian matrix. Furthermore, we discuss the case of time-varying reference vectors and provide a convergence result for this case. The latter helps reach synchronization, requiring less communication links and weaker conditions on the instantaneous reference vectors. Our methods are illustrated on an attitude synchronization problem where agents exchange only their relative positions observed in the respective body frames. ©2009 IEEE.

Stabilization of symmetric formations to motion around convex loops

We provide a cooperative control algorithm to stabilize symmetric formations to motion around closed curves suitable for mobile sensor networks. This work extends previous results for stabilization of symmetric circular formations. We study a planar particle model with decentralized steering control subject to limited communication. Because of their unique spectral properties, the Laplacian matrices of circulant graphs play a key role. We illustrate the result for a skewed superellipse, which is a type of curve that includes circles, ellipses, and rounded parallelograms. © 2007 Elsevier B.V. All rights reserved.

Collective motion of self-propelled particles: Stabilizing symmetric formations on closed curves

We provide feedback control laws to stabilize formations of multiple, unit speed particles on smooth, convex, and closed curves with definite curvature. As in previous work we exploit an analogy with coupled phase oscillators to provide controls which isolate symmetric particle formations that are invariant to rigid translation of all the particles. In this work, we do not require all particles to be able to communicate; rather we assume that inter-particle communication is limited and can be modeled by a fixed, connected, and undirected graph. Because of their unique spectral properties, the Laplacian matrices of circulant graphs play a key role. The methodology is demonstrated using a superellipse, which is a type of curve that includes circles, ellipses, and rounded rectangles. These results can be used in applications involving multiple autonomous vehicles that travel at constant speed around fixed beacons. ©2006 IEEE.

Smoothing LDA Model for Text Categorization

Abstract. Latent Dirichlet Allocation (LDA) is a document level language model. In general, LDA employ the symmetry Dirichlet distribution as prior of the topic-words’ distributions to implement model smoothing. In this paper, we propose a data-driven smoothing strategy in which probability mass is allocated from smoothing-data to latent variables by the intrinsic inference procedure of LDA. In such a way, the arbitrariness of choosing latent variables'priors for the multi-level graphical model is overcome. Following this data-driven strategy,two concrete methods, Laplacian smoothing and Jelinek-Mercer smoothing, are employed to LDA model. Evaluations on different text categorization collections show data-driven smoothing can significantly improve the performance in balanced and unbalanced corpora.

Inner structure of Spin(c)(4) gauge potential on 4-dimensional manifolds

The decomposition of Spin(c)(4) gauge potential in terms of the Dirac 4-spinor is investigated, where an important characterizing equation Delta A(mu) = -lambda A(mu) has been discovered. Here, lambda is the vacuum expectation value of the spinor field, lambda = parallel to Phi parallel to(2), and A(mu) the twisting U(1) potential. It is found that when), takes constant values, the characterizing equation becomes an eigenvalue problem of the Laplacian operator. It provides a revenue to determine the modulus of the spinor field by using the Laplacian spectral theory. The above study could be useful in determining the spinor field and twisting potential in the Seiberg-Witten equations. Moreover, topological characteristic numbers of instantons in the self-dual sub-space are also discussed.

基于隐含狄利克雷分配模型的文本分类方法研究

随着互联网的不断发展，网络文本信息资源正在急剧增加，如何利用计算机自动有效地对海量的文本信息进行管理和查询成为了当前的迫切需求。而文本分类技术提供了一种对自然语言文档进行自动组织的有效手段。本文从模型平滑、类别信息嵌入和高性能求解算法等方面对LDA 模型（Latent Dirichlet Allocation，隐含狄利克雷分配）开展了研究，研究内容涉及文本分类中不平衡语料分类、文本表示、复杂分类器加速等多个方面，论文的主要工作和创新点总结如下：第一、为了克服传统LDA 模型平滑算法中直接修改多级图模型隐含变量分布的随意性我们提出了数据驱动的Laplacian 平滑方法和数据驱动的 Jelinek-Mercer 平滑方法。数据驱动的Laplacian 平滑方法缓解了传统LDA 模型的过拟合现象，数据驱动的Jelinek-Mercer 平滑方法在基本保持整体性能的情况下，降低了预测阶段的时间复杂度。数据驱动的平滑策略在平衡和非平衡语料库上都能够显著提高LDA 模型的分类性能。第二、传统LDA 模型计算目标文档在不同类别上的生成概率时，在文档非所属类别的隐含主题上会发生强制分配。针对这一问题我们提出了Labeled-LDA 模型将隐含主题和类别信息融合在一起，在分类目标文档时协同计算全部类别的隐含主题的分配以改善分类性能。第三、LDA 等隐含主题模型是近期在文本挖掘领域发展起来的重要研究方向，但是算法具有较高的计算复杂度。我们以动态负载均衡的算法为重点，研究和实现了LDA 模型在多内核计算机上的并行计算和CTM 模型（Correlated Topic Model，关联主题模型）在异构集群环境下的分布式计算。第四、基于提出的Labeled-LDA 模型我们设计和实现了一个具有柔性文本分类功能的实验系统。该系统利用Labeled-LDA 模型对目标文档中的隐含主题进行概率推断，进而得到文档在各类别上的分配量。相对概率支持向量机等判别模型概率估计方法，该分配量具有较明确的实际意义，同时避免了贝叶斯等生成分类模型输出极端概率值的问题。