Smoothness in Bayesian Non-parametric Regression with Wavelets

Consider the classical nonparametric regression problem y_i = f(t_i) + _ii = 1,...,n where t_i = i/n, and _i are i.i.d. zero mean normal with variance ². The aim is to estimate the true function f which is assumed to belong to the smoothness class described by the Besov space B _pq ^q . These are functions belonging to L_p with derivatives up to order s, in L_p sense. The parameter q controls a further finer degree of smoothness. In a Bayesian setting, a prior on B _pq ^q is chosen following Abramovich, Sapatinas and Silverman (1998). We show that the optimal Bayesian estimator of f is then also a.s. in B _pq ^q if the loss function is chosen to be the Besov norm of B _pq ^q . Because it is impossible to compute this optimal Bayesian estimator analytically, we propose a stochastic algorithm based on an approximation of the Bayesian risk and simulated annealing. Some simulations are presented to show that the algorithm performs well and that the new estimator is competitive when compared to the more standard posterior mean.     

Energy as a Fifth Dimension

A recent analysis of the experimental data on some physical phenomena ruled by the four fundamental interactions (electromagnetic, weak, strong and gravitational) seems to show the possibility of describing such interactions in terms of a deformation of the usual Minkowski spacetime, with a metric whose coefficients do depend on the energy of the process considered. In this paper, we show that such results can be accounted for in terms of a Kaluza-Klein-like scheme, based on a five-dimensional Riemannian space in which energy plays the role of the fifth dimension. The corresponding five-dimensional Einstein equations in vacuum are solved in some cases of physical relevance and it is shown that all the phenomenological metrics describing the four fundamental forces are recovered as special cases of the classes of solutions found. Possible developments of the formalism are also briefly outlined.     

Personal neutron dosimetry in the space station MIR and the Space Shuttle

A passive neutron dosemeter based on nuclear track detectors and TLD's was used in 1995 and 1997 on the MIR station and in Space Shuttle flights to MIR. As it is equipped with neutron converters and shieldings of different types the track detector system allows the neutron dose equivalent to be determined in rough energy intervals. The results of the measurements on the MIR station and in the Space Shuttle flights are presented and the influence of charged particles in the complex mixed radiation field in space is discussed. Improvements are possible by means of a new active neutron dosemeter which is under development at the PTB. First measurements with a prototype in the high-energy reference fields at CERN are presented and discussed.     

线性空间中与子空间的并集合基本上不相交的子空间的存在性及维数

在线性空间V中,对于一些真子空间的并集合M来说,一定存在着V的非零子空间V0使得M∩V0={0},并且这些V0的最大维数可确定。     

Fuzzy度量空间中映像族的公共不动点定理

本文讨论了Fuzzy度量空间中映像对和映像族的公共不动点问题,所获定理推广了Kaleva.O Seikkala. S和郑叔的有关结果。     

Banach空间中混合型非线性积分──微分方程的最大解和最小解

在弱序列完备的Ｂａｎａｃｈ空间中，利用推理论和单调迭代方法，研究了混合型非线性积分－微分方程初值问题最大解和最小解的存在性。其结果改进和推广了许多已知的结果。     

四维复空间的超复数和单位球

利用克里复德式数在四维复空间中引入两类超复数。讨论两类复空间的几何性质,用四维球谐函数推导出单位园的体积及面积的计算公式。     

空间权法律问题浅析

随着城市人口的快速增长和土地资源的日趋稀缺,以立体方式利用土地,已成为世界范围的普遍趋势。空间权是人类对土地利用由平面向立体形态发展的一种新型权利形式。与西方各国比较而言,我国无论在土地立体利用的实践层面,还是在立法层面,都处于相对落后状态。鉴于“立体的土地立法”已成为土地立法的基本趋势,我国应借鉴西方各国及其他地区的立法经验,在我国的物权法中确立空间权制度。     

轨道空间上的运费不等式

研究了一类概率距离dp，它是全变差范数的自然推广．为方便运用这一抽象的距离，利用Wasserstein耦合方法，在Polish空间上用相对熵给出dp的上界．最后将这类距离推广到轨道空间上，建立了一类运费不等式．     

拟相似算子近似点谱的连通分支

讨论了Banach空间X上两个算子T，S拟相似时，近似点谱σα(T)的每一个连通分支与σα（S）以及σs(S)的相交关系，证明了σα(T)的每一个连通分支与σs(S)的交非空，并且给出了σα（T）的连通分支与σα（S）交非空的充要条件。