Quantum graphs attracted a lot of interest recently. There are several reasons for that. On one hand these models are useful as descriptions of various structures prepared from semiconductor wires, carbon nanotubes, and other substances. On the other hand they provide a tool to study properties of quantum dynamics in situations when the system has a nontrivial geometrical or topological structure. Quantum graph models contain typically free parameters related to coupling of the wave functions at the graph vertices, and to get full grasp of the theory one has to understand their physical meaning. A natural approach to this question is to investigate fat graphs, that is, systems of thin tubes built over the skeleton of a given graph, and to analyze its limit as the tube thickness tends to zero.

We introduce an irrational factor of order <i>k</i> defined by I k ( n ) = i = 1 l p i i , where I k ( n ) = i = 1 l p i i is the factorization of <i>n</i> and $${\beta_{i} = \left\{\begin{array}{ll}\alpha_i, \quad \quad {\rm if} \quad \alpha_i &lt; k \\ \frac{1}{\alpha_i},\quad \quad {\rm if} \quad \alpha_i \geqq k \end{array}\right.}$$. It turns out that the function I k ( n ) = i = 1 l p i i well approximates the characteristic function of <i>k</i>-free integers. We also derive asymptotic formulas for I k ( n ) = i = 1 l p i i and I k ( n ) = i = 1 l p i i .

Kawakubo and Uchida showed that, if a closed oriented 4k-dimensional manifold M admits a semi-free circle action such that the dimension of the fixed point set is less than 2k, then the signature of M vanishes. In this note, by using G-signature theorem and the rigidity of the signature operator, we generalize this result to more general circle actions. Combining the same idea with the remarkable WittenTaubesBott rigidity theorem, we explore more vanishing results on spin manifolds admitting such circle actions. Our results are closely related to some earlier results of ConnerFloyd, LandweberStong and HirzebruchSlodowy.

Consider a linear model Y= X + z, where X= X n, p and z~ N (0, I n). The vector is unknown and it is of interest to separate its nonzero coordinates from the zero ones (ie, variable selection). Motivated by examples in long-memory time series (Fan and Yao, 2003) and the change-point problem (Bhattacharya, 1994), we are primarily interested in the case where the Gram matrix G= X X is non-sparse but sparsifiable by a finite order linear filter. We focus on the regime where signals are both rare and weak so that successful variable selection is very challenging but is still possible.

In their proposed compactification of superstrings [4], Candelas, Horowitz, Strominger and Witten took the matric product of a maximal symmetric four dimensional spacetime M with a six dimensional CalabiYau vacua X as the ten dimensional spacetime; they identified the YangMills connection with the SU (3) connection of the CalabiYau metric and set the dilaton to be a constant. To make this theory compatible with the standard grand unified field theory, Witten [28] and HoravaWitten [20] proposed to use higher rank bundles for strong coupled heterotic string theory so that the gauge groups can be SU (4) or SU (5). Mathematically, this approach relies on UhlenbeckYaus theorem on constructing HermitianYangMills connections on stable bundles [27]. Many authors, including Friedman, Morgan and Witten [18]; Donagi, Ovrat, Pantev and Reinbacher [12]; Andreas [1], Kachru [21] and others, have worked