首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 453 毫秒
1.
2.
In many applications it has been observed that hybrid-Monte Carlo sequences perform better than Monte Carlo and quasi-Monte Carlo sequences, especially in difficult problems. For a mixed ss-dimensional sequence mm, whose elements are vectors obtained by concatenating dd-dimensional vectors from a low-discrepancy sequence qq with (s−d)(sd)-dimensional random vectors, probabilistic upper bounds for its star discrepancy have been provided. In a paper of G. Ökten, B. Tuffin and V. Burago [G. Ökten, B. Tuffin, V. Burago, J. Complexity 22 (2006), 435–458] it was shown that for arbitrary ε>0ε>0 the difference of the star discrepancies of the first NN points of mm and qq is bounded by εε with probability at least 1−2exp(−ε2N/2)12exp(ε2N/2) for NN sufficiently large. The authors did not study how large NN actually has to be and if and how this actually depends on the parameters ss and εε. In this note we derive a lower bound for NN, which significantly depends on ss and εε. Furthermore, we provide a probabilistic bound for the difference of the star discrepancies of the first NN points of mm and qq, which holds without any restrictions on NN. In this sense it improves on the bound of Ökten, Tuffin and Burago and is more helpful in practice, especially for small sample sizes NN. We compare this bound to other known bounds.  相似文献   

3.
The paper deals with the radially symmetric solutions of ut=Δu+um(x,t)vn(0,t)ut=Δu+um(x,t)vn(0,t), vt=Δv+up(0,t)vq(x,t)vt=Δv+up(0,t)vq(x,t), subject to null Dirichlet boundary conditions. For the blow-up classical solutions, we propose the critical exponents for non-simultaneous blow-up by determining the complete and optimal classification for all the non-negative exponents: (i) There exist initial data such that uu (vv) blows up alone if and only if m>p+1m>p+1 (q>n+1q>n+1), which means that any blow-up is simultaneous if and only if m≤p+1mp+1, q≤n+1qn+1. (ii) Any blow-up is uu (vv) blowing up with vv (uu) remaining bounded if and only if m>p+1m>p+1, q≤n+1qn+1 (m≤p+1mp+1, q>n+1q>n+1). (iii) Both non-simultaneous and simultaneous blow-up may occur if and only if m>p+1m>p+1, q>n+1q>n+1. Moreover, we consider the blow-up rate and set estimates which were not obtained in the previously known work for the same model.  相似文献   

4.
5.
An automatic quadrature method is presented for approximating fractional derivative Dqf(x)Dqf(x) of a given function f(x)f(x), which is defined by an indefinite integral involving f(x)f(x). The present method interpolates f(x)f(x) in terms of the Chebyshev polynomials in the range [0, 1] to approximate the fractional derivative Dqf(x)Dqf(x) uniformly for 0≤x≤10x1, namely the error is bounded independently of xx. Some numerical examples demonstrate the performance of the present automatic method.  相似文献   

6.
Let KK be a closed convex subset of a qq-uniformly smooth separable Banach space, T:K→KT:KK a strictly pseudocontractive mapping, and f:K→Kf:KK an LL-Lispschitzian strongly pseudocontractive mapping. For any t∈(0,1)t(0,1), let xtxt be the unique fixed point of tf+(1-t)Ttf+(1-t)T. We prove that if TT has a fixed point, then {xt}{xt} converges to a fixed point of TT as tt approaches to 0.  相似文献   

7.
We prove that if for a continuous map ff on a compact metric space XX, the chain recurrent set, R(f)R(f) has more than one chain component, then ff does not satisfy the asymptotic average shadowing property. We also show that if a continuous map ff on a compact metric space XX has the asymptotic average shadowing property and if AA is an attractor for ff, then AA is the single attractor for ff and we have A=R(f)A=R(f). We also study diffeomorphisms with asymptotic average shadowing property and prove that if MM is a compact manifold which is not finite with dimM=2dimM=2, then the C1C1 interior of the set of all C1C1 diffeomorphisms with the asymptotic average shadowing property is characterized by the set of ΩΩ-stable diffeomorphisms.  相似文献   

8.
The Feedback Vertex Set problem asks whether a graph contains qq vertices meeting all its cycles. This is not a local property, in the sense that we cannot check if qq vertices meet all cycles by looking only at their neighbors. Dynamic programming algorithms for problems based on non-local properties are usually more complicated. In this paper, given a graph GG of clique-width cwcw and a cwcw-expression of GG, we solve the Minimum Feedback Vertex Set problem in time O(n22O(cwlogcw))O(n22O(cwlogcw)). Our algorithm applies dynamic programming on a so-called kk-module decomposition of a graph, as defined by Rao (2008) [29], which is easily derivable from akk-expression of the graph. The related notion of module-width of a graph is tightly linked to both clique-width and NLC-width, and in this paper we give an alternative equivalent characterization of module-width.  相似文献   

9.
Let RR be a commutative ring with identity. We will say that an RR-module MM satisfies the weak Nakayama property, if IM=MIM=M, where II is an ideal of RR, implies that for any x∈MxM there exists a∈IaI such that (a−1)x=0(a1)x=0. In this paper, we will study modules satisfying the weak Nakayama property. It is proved that if RR is a local ring, then RR is a Max ring if and only if J(R)J(R), the Jacobson radical of RR, is TT-nilpotent if and only if every RR-module satisfies the weak Nakayama property.  相似文献   

10.
It is proved that the solutions to the singular stochastic pp-Laplace equation, p∈(1,2)p(1,2) and the solutions to the stochastic fast diffusion equation with nonlinearity parameter r∈(0,1)r(0,1) on a bounded open domain Λ⊂RdΛRd with Dirichlet boundary conditions are continuous in mean, uniformly in time, with respect to the parameters pp and rr respectively (in the Hilbert spaces L2(Λ)L2(Λ), H−1(Λ)H1(Λ) respectively). The highly singular limit case p=1p=1 is treated with the help of stochastic evolution variational inequalities, where PP-a.s. convergence, uniformly in time, is established.  相似文献   

11.
12.
This paper is concerned with the Cauchy problem for the fast diffusion equation ut−Δum=αup1utΔum=αup1 in RNRN (N≥1N1), where m∈(0,1)m(0,1), p1>1p1>1 and α>0α>0. The initial condition u0u0 is assumed to be continuous, nonnegative and bounded. Using a technique of subsolutions, we set up sufficient conditions on the initial value u0u0 so that u(t,x)u(t,x) blows up in finite time, and we show how to get estimates on the profile of u(t,x)u(t,x) for small enough values of t>0t>0.  相似文献   

13.
Suppose XX is a real qq-uniformly smooth Banach space and F,K:X→XF,K:XX are Lipschitz ??-strongly accretive maps with D(K)=F(X)=XD(K)=F(X)=X. Let uu denote the unique solution of the Hammerstein equation u+KFu=0u+KFu=0. An iteration process recently introduced by Chidume and Zegeye is shown to converge strongly to uu. No invertibility assumption is imposed on KK and the operators KK and FF need not be defined on compact subsets of XX. Furthermore, our new technique of proof is of independent interest. Finally, some interesting open questions are included.  相似文献   

14.
Suppose XX is a real qq-uniformly smooth Banach space and F,K:X→XF,K:XX are bounded strongly accretive maps with D(K)=F(X)=XD(K)=F(X)=X. Let uu denote the unique solution of the Hammerstein equation u+KFu=0u+KFu=0. A new explicit coupled iteration process is shown to converge strongly to uu. No invertibility assumption is imposed on KK and the operators KK and FF need not be defined on compact subsets of XX. Furthermore, our new technique of proof is of independent interest. Finally, some interesting open questions are included.  相似文献   

15.
We introduce (n+1)(n+1)-preprojective algebras of algebras of global dimension nn. We show that if an algebra is nn-representation-finite then its (n+1)(n+1)-preprojective algebra is self-injective. In this situation, we show that the stable module category of the (n+1)(n+1)-preprojective algebra is (n+1)(n+1)-Calabi–Yau, and, more precisely, it is the (n+1)(n+1)-Amiot cluster category of the stable nn-Auslander algebra of the original algebra. In particular this stable category contains an (n+1)(n+1)-cluster tilting object. We show that even if the (n+1)(n+1)-preprojective algebra is not self-injective, under certain assumptions (which are always satisfied for n∈{1,2}n{1,2}) the results above still hold for the stable category of Cohen–Macaulay modules.  相似文献   

16.
In this paper, we consider Beta(2−α,α)(2α,α) (with 1<α<21<α<2) and related ΛΛ-coalescents. If T(n)T(n) denotes the length of a randomly chosen external branch of the nn-coalescent, we prove the convergence of nα−1T(n)nα1T(n) when nn tends to ∞, and give the limit. To this aim, we give asymptotics for the number σ(n)σ(n) of collisions which occur in the nn-coalescent until the end of the chosen external branch, and for the block counting process associated with the nn-coalescent.  相似文献   

17.
Let x(s)x(s), s∈RdsRd be a Gaussian self-similar random process of index HH. We consider the problem of log-asymptotics for the probability pTpT that x(s)x(s), x(0)=0x(0)=0 does not exceed a fixed level in a star-shaped expanding domain T⋅ΔTΔ as T→∞T. We solve the problem of the existence of the limit, θ?lim(−logpT)/(logT)Dθ?lim(logpT)/(logT)D, T→∞T, for the fractional Brownian sheet x(s)x(s), s∈[0,T]2s[0,T]2 when D=2D=2, and we estimate θθ for the integrated fractional Brownian motion when D=1D=1.  相似文献   

18.
Let us fix a function f(n)=o(nlnn)f(n)=o(nlnn) and real numbers 0≤α<β≤10α<β1. We present a polynomial time algorithm which, given a directed graph GG with nn vertices, decides either that one can add at most βnβn new edges to GG so that GG acquires a Hamiltonian circuit or that one cannot add αnαn or fewer new edges to GG so that GG acquires at least e−f(n)n!ef(n)n! Hamiltonian circuits, or both.  相似文献   

19.
We consider a multidimensional diffusion XX with drift coefficient b(α,Xt)b(α,Xt) and diffusion coefficient ?σ(β,Xt)?σ(β,Xt). The diffusion sample path is discretely observed at times tk=kΔtk=kΔ for k=1…nk=1n on a fixed interval [0,T][0,T]. We study minimum contrast estimators derived from the Gaussian process approximating XX for small ??. We obtain consistent and asymptotically normal estimators of αα for fixed ΔΔ and ?→0?0 and of (α,β)(α,β) for Δ→0Δ0 and ?→0?0 without any condition linking ?? and ΔΔ. We compare the estimators obtained with various methods and for various magnitudes of ΔΔ and ?? based on simulation studies. Finally, we investigate the interest of using such methods in an epidemiological framework.  相似文献   

20.
A plane partition   is a p×qp×q matrix A=(aij)A=(aij), where 1?i?p1?i?p and 1?j?q1?j?q, with non-negative integer entries, and whose rows and columns are weakly decreasing. From a geometric point of view plane partitions are equivalent to pyramids  , subsets of the integer lattice Z3Z3 which play an important role in Discrete Tomography. As a consequence, some typical problems concerning the tomography of discrete lattice sets can be rephrased and considered via plane partitions. In this paper we focus on some of them. In particular, we get a necessary and sufficient condition for additivity, a canonical procedure for checking the existence of (weakly) bad configurations, and an algorithm which constructs minimal pyramids (with respect to the number of levels) with assigned projection of a bad configurations.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号