Conformally invariant field theories in two dimensions and corresponding statistical mechanical models

Belavin, Zamlodochikov and Polyakov have recently proposed a class of conformally invariant field theories in two dimension with exactly determined rational critical indices. We establish a tentative identification of a subset of these theories in terms of the O(n) model and theq-state Potts model in 2-dimensions for appropriaten andq. The results of this work were reported in the conference on “Structural Similarities in Exactly Solved Models” at I.T.P. Santa Barbara, August 1984.     

Weak convergence in applied probability

Weak convergence of probability measures on function spaces has been active area of research in recent years. While the theory has a somewhat abstract base, it is extremely useful in a wide variety of problems and we believe has much to offer to applied probability. Our aim in this survey paper is to discuss those aspects of the theory which are relevant to work in applied probability. After an introduction to the foundations of weak convergence, we shall discuss partial sum, point, Markov and extremal processes. These processes form the building blocks for many of the important models of applied probability.     

Asymptotic Stability of Transiently Chaotic Neural Networks

The asymptotic stability of transiently chaotic neural networks is considered in this paper. By utilizing a Lyapunov function and then applying LaSalle's invariance principle, we present several sufficient conditions that guarantee the stability. These results are less conservative than those established in the earlier references. Two numerical examples are given to illustrate the applicability of these conditions.     

Association Schemes and Fusion Algebras (An Introduction)

We introduce the concept of fusion algebras at algebraic level, as a purely algebraic concept for the fusion algebras which appear in conformal field theory in mathematical physics. We first discuss the connection between fusion algebras at algebraic level and character algebras, a purely algebraic concept for Bose-Mesner algebras of association schemes. Through this correspondence, we establish the condition when the matrix S of a fusion algebra at algebraic level is unitary or symmetric. We construct integral fusion algebras at algebraic level, from association schemes, in particular from group association schemes, whose matrix S is unitary and symmetric. Finally, we consider whether the modular invariance property is satisfied or not, namely whether there exists a diagonal matrix T satisfying the condition (ST)³ = S ². We prove that this property does not hold for some integral fusion algebras at algebraic level coming from the group association scheme of certain groups of order 64, and we also prove that the (nonintegral) fusion algebra at algebraic level obtained from the Hamming association scheme H(d, q) has the modular invariance property.     

基于方向-频率分解的旋转不变性纹理分类

提出了一种用于纹理分类的旋转不变性特征提取的新算法.该算法是将一定大小的图像进行二维傅里叶变换;其次在变换后的图像中央选择一个圆盘区域,并在方向[0°,180°]内进行等间隔角度频率抽样,实现方向分解,使用一组复Morlet小波对每个方向上的映射切片进行小波变换,从而实现多通道频率分解;在各个频率通道中计算均值和方差作为特征,并利用线性回归模型计算频率通道之间的关系特征;将特征沿方向进行一维傅里叶变换并取其幅值,从而得到旋转不变性特征.实验结果表明所提取的特征具有较好的旋转不变性,与其它算法相比具有更好的分类性能,并且对无旋转纹理分类也能产生较好的分类结果.     

The Averaged Robbins – Monro Method for Linear Problems in a Banach Space

We consider a recursive method of Robbins–Monro type to solve the linear problem Ax=V in a Banach space. The bounded linear operator A and the vector V are assumed to be observable with some noise only. According to Polyak and Ruppert we use gains converging to zero slower than 1/n and take the average of the iterates as an estimator for the solution of the linear problem. Under weak conditions on the noise processes almost sure and distributional invariance principles are shown.     

Why is \mathcal{CPT} Fundamental?

Lüders and Pauli proved the theorem based on Lagrangian quantum field theory almost half a century ago. Jost gave a more general proof based on “axiomatic” field theory nearly as long ago. The axiomatic point of view has two advantages over the Lagrangian one. First, the axiomatic point of view makes clear why is fundamental—because it is intimately related to Lorentz invariance. Secondly, the axiomatic proof gives a simple way to calculate the transform of any relativistic field without calculating , and separately and then multiplying them. The purpose of this pedagogical paper is to “deaxiomatize” the theorem by explaining it in a few simple steps. We use theorems of distribution theory and of several complex variables without proof to make the exposition elementary.     

The regulator problem for linear systems with constrained control: an LMI approach

^** Email: benzaouia{at}ucam.ac.ma A new methodology of the partial eigenstructure assignment bystate feedback via linear matrix inequality (LMI) is extendedto obtain a solution of the constrained regulator problem forlinear continuous-time and discrete-time systems by using theLMI formulation.