A safeguarded dual weighted residual method

Andreas Veeser ^{The dual weighted residual (DWR) method yields reliable a posteriorierror bounds for linear output functionals provided that theerror incurred by the numerical approximation of the dual solutionis negligible. In that case, its performance is generally superiorthan that of global ‘energy norm’ error estimatorswhich are ‘unconditionally’ reliable. We presenta simple numerical example for which neglecting the approximationerror leads to severe underestimation of the functional error,thus showing that the DWR method may be unreliable. We proposea remedy that preserves the original performance, namely a DWRmethod safeguarded by additional asymptotically higher ordera posteriori terms. In particular, the enhanced estimator isunconditionally reliable and asymptotically coincides with theoriginal DWR method. These properties are illustrated via theaforementioned example.}     

A uniformly stable family of mixed hp-finite elements with continuous pressures for incompressible flow

A new family of mixed hp-finite elements is presented for thediscretization of planar Stokes flow on meshes of curvilinear,quadrilateral elements. The elements involve continuous pressuresand are shown to be stable with an inf–sup constant boundedbelow independently of the mesh-size h and the spectral orderp. The spaces have balanced approximation properties—theorders of approximation in h and p are equal for both the velocityand the pressure. This is the first example of a uniformly stablemethod with continuous pressures for spectral element discretizationof Stokes equations, valid for geometrically refined meshesand curvilinear elements.     

A posteriori L2 error estimation on anisotropic tetrahedral finite element meshes

A new a posteriori L₂ norm error estimator is proposed for thePoisson equation. The error estimator can be applied to anisotropictetrahedral or triangular finite element meshes. The estimatoris rigorously analysed for Dirichlet and Neumann boundary conditions. The lower error bound relies on specifically designed anisotropicbubble functions and the corresponding inverse inequalities.The upper error bound utilizes non-standard anisotropic interpolationestimates. Its proof requires H² regularity of the Poisson problem,and its quality depends on how good the anisotropic mesh resolvesthe anisotropy of the problem. This is measured by a so-called‘matching function’. A numerical example supports the anisotropic error analysis.     

Products of Consecutive Integers

In this paper, a number of results are deduced on the arithmeticstructure of products of integers in short intervals. By wayof an example, work of Saradha and Hanrot, and of Saradha andShorey, is completed by the provision of an answer to the questionof when the product of k out of k + 1 consecutive positive integerscan be an ‘almost’ perfect power. The main new ingredientin these proofs is what might be termed a practical method forresolving high-degree binomial Thue equations of the form axⁿ–byⁿ= ±1, based upon results from the theory of Galois representationsand modular forms. 2000 Mathematics Subject Classification 11D41,11D61.     

Dual Pairs of Hopf *-Algebras

If A is a Hopf *-algebra, the dual space A' is again a *-algebra.There is a natural subalgebra A° of A' that is again a Hopf*-algebra. In many interesting examples, A° will be largeenough (to separate points of A). More generally, one can considera pair (A, B) of Hopf *-algebras and a bilinear form on A xB with conditions such that, if the pairing is non-degenerate,one algebra can be considered as a subalgebra of the dual ofthe other. In these notes, we study such pairs of Hopf *-algebras. We startfrom the notion of a Hopf *-algebra A and its reduced dual A°.We give examples of pairs of Hopf *-algebras, and discuss theproblem of non-degeneracy. The first example is an algebra pairedwith itself. The second example is the pairing of a Hopf *-algebra(due to Jimbo) and the twisted SU(n) of Woronowicz. We alsodiscuss the notion of the quantum double of Drinfeld in thisframework of dual pairs.     

Modular Form Congruences and Selmer Groups

Motivated by Cremona and Mazur's notion of visibility of elementsin Shafarevich–Tate groups [6, 27], there have been anumber of recent works which test its compatibility with theBirch and Swinnerton–Dyer conjecture and the Bloch–Katoconjecture. These conjectures provide formulas for the ordersof Shafarevich–Tate groups in terms of values of L-functions.For example, one may see recent work of Agashe, Dummigan, Steinand Watkins [1, 2, 10, 11]. In their examples, they find thatthe presence of visible elements agrees with the expected divisibilityproperties of the relevant L-values.     

Logarithmic Growth for Matrix Martingale Transforms

An example is given of an operator weight W that satisfies thedyadic operator Hunt–Muckenhoupt–Wheeden condition for which there exists a dyadic martingale transform on L² (W) that is unbounded. The constructionrelates weighted boundedness to the boundedness of dyadic vectorHankel operators.     

Operator amenability of Fourier-Stieltjes algebras, II

We give an example of a non-compact, locally compact group Gsuch that its Fourier–Stieltjes algebra B (G) is operatoramenable. Furthermore, we characterize those G for which A *(G),the spine of B (G) as introduced by M. Ilie and N. Spronk, isoperator amenable and show that A *(G) is operator weakly amenablefor each G.     

Jaffard-Ohm correspondence and Hochster duality

We study categorical aspects of the Jaffard–Ohm correspondencebetween abelian l-groups and Bézout domains and showthat this correspondence is close to a localization. For thispurpose, we establish a general extension theorem for valuationswith value group that is an abelian l-group. As an application,we prove Anderson's conjecture which refines the Jaffard–Ohmcorrespondence. We then extend the correspondence to sheaveson spectral spaces and show that the spectrum of a Bézoutdomain and the spectrum of its corresponding abelian l-groupprovide a concrete example for Hochster's duality of spectralspaces.     

Multiple Positive Solutions of Semilinear Differential Equations with Singularities

The existence of positive solutions of a second order differentialequation of the form z'+g(t)f(z)=0 (1.1) with the separated boundary conditions: z(0) – ßz'(0)= 0 and z(1)+z'(1) = 0 has proved to be important in physicsand applied mathematics. For example, the Thomas–Fermiequation, where f = z^3/2 and g = t^–1/2 (see [12, 13, 24]),so g has a singularity at 0, was developed in studies of atomicstructures (see for example, [24]) and atomic calculations [6].The separated boundary conditions are obtained from the usualThomas–Fermi boundary conditions by a change of variableand a normalization (see [22, 24]). The generalized Emden–Fowlerequation, where f = z^p, p > 0 and g is continuous (see [24,28]) arises in the fields of gas dynamics, nuclear physics,chemically reacting systems [28] and in the study of multipoletoroidal plasmas [4]. In most of these applications, the physicalinterest lies in the existence and uniqueness of positive solutions.     

Cohen-Macaulay Complexes and Koszul Rings

Throughout this paper k denotes a fixed commutative ground ring.A Cohen–Macaulay complex is a finite simplicial complexsatisfying a certain homological vanishing condition. Thesecomplexes have been the subject of much research; introductionscan be found in, for example, Björner, Garsia and Stanley[6] or Budach, Graw, Meinel and Waack [7]. It is known (see,for example, Cibils [8], Gerstenhaber and Schack [10]) thatthere is a strong connection between the (co)homology of anarbitrary simplicial complex and that of its incidence algebra.We show how the Cohen–Macaulay property fits into thispicture, establishing the following characterization. A pure finite simplicial complex is Cohen–Macaulay overk if and only if the incidence algebra over k of its augmentedface poset, graded in the obvious way by chain lengths, is aKoszul ring.     

Zero-Sets of Quaternionic and Octonionic Analytic Functions with Central Coefficients

We prove that the zero set of any quaternionic (or octonionic)analytic function f with central (that is, real) coefficientsis the disjoint union of codimension two spheres in R⁴ or R⁸(respectively) and certain purely real points. In particular,for polynomials with real coefficients, the complete root-setis geometrically characterisable from the lay-out of the rootsin the complex plane. The root-set becomes the union of a finitenumber of codimension 2 Euclidean spheres together with a finitenumber of real points. We also find the preimages f^–1for any quaternion (or octonion) A. We demonstrate that this surprising phenomenon of complete spheresbeing part of the solution set is very markedly a special ‘real’phenomenon. For example, the quaternionic or octonionic Nthroots of any non-real quaternion (respectively octonion) turnout to be precisely N distinct points. All this allows us todo some interesting topology for self-maps of spheres.     

Logarithmic Orbifold Euler Numbers of Surfaces with Applications

We introduce orbifold Euler numbers for normal surfaces withboundary Q-divisors. These numbers behave multiplicatively underfinite maps and in the log canonical case we prove that theysatisfy the Bogomolov–Miyaoka–Yau type inequality.Existence of such a generalization was earlier conjectured byG. Megyesi [Proc. London Math. Soc. (3) 78 (1999) 241–282].Most of the paper is devoted to properties of local orbifoldEuler numbers and to their computation. As a first application we show that our results imply a generalizedversion of R. Holzapfel's ‘proportionality theorem’[Ball and surface arithmetics, Aspects of Mathematics E29 (Vieweg,Braunschweig, 1998)]. Then we show a simple proof of a necessarycondition for the logarithmic comparison theorem which recoversan earlier result by F. Calderón-Moreno, F. Castro-Jiménez,D. Mond and L. Narváez-Macarro [Comment. Math. Helv.77 (2002) 24–38]. Then we prove effective versions of Bogomolov's result on boundednessof rational curves in some surfaces of general type (conjecturedby G. Tian [Springer Lecture Notes in Mathematics 1646 (1996)143–185)]. Finally, we give some applications to singularitiesof plane curves; for example, we improve F. Hirzebruch's boundon the maximal number of cusps of a plane curve. 2000 MathematicalSubject Classification: 14J17, 14J29, 14C17.     

Reduced-order filtering with energy-to-peak performance for discrete-time Markovian jumping systems

In this paper, the l₂–l (energy-to-peak) performanceof the discrete-time Markovian jump linear system is investigated.The jump parameters are modelled by a discrete-time Markov process.Furthermore, we study the l₂–l reduced-order filteringproblem for the Markovian jump linear system. A reduced-orderfilter with the same randomly jumping parameters is proposedwhich can make the error systems with Markovian jump parametersstochastically stable with a prescribed l₂–lperformance.Sufficient conditions in terms of linear matrix inequalities(LMIs) and a coupling non-convex rank constraint are derivedfor the existence of a solution to the reduced-order filteringproblems. A numerical example is given to illustrate the designprocedures.     

Kazhdan-Lusztig Cells, q-Schur Algebras and James' Conjecture

We consider the Dipper–James q-Schur algebra S_q(n, r)_k,defined over a field k and with parameter q 0. An understandingof the representation theory of this algebra is of considerableinterest in the representation theory of finite groups of Lietype and quantum groups; see, for example, [6] and [11]. Itis known that S_q(n, r)_k is a semisimple algebra if q is nota root of unity. Much more interesting is the case when S_q(n,r)_k is not semisimple. Then we have a corresponding decompositionmatrix which records the multiplicities of the simple modulesin certain ‘standard modules’ (or ‘Weyl modules’).A major unsolved problem is the explicit determination of thesedecomposition matrices.     

Solution to a Class of Coupled Linear Partial Differential Equations

The solution to a coupled system of partial differential equationsinvolving a general linear time-independent operator L is presented.Examples of these equations include coupled diffusion equationsor coupled convection–dispersion equations. The solutionconsists of a convolution of the Green's function appropriatefor the operator L and a function independent of the operatorL. The method enables one to write software to calculate thesolution to a wide range of problems. The change of solutionupon changing the problem often only involves a substitutionof the Green's function. A specific example of physical significanceis given.     

Splendid Derived Equivalences for Blocks of Finite Groups

A central issue in finite group modular representation theoryis the relationship between the p-local structure and the p-modularrepresentation theory of a given finite group. In [5], Brouéposes some startling conjectures. For example, he conjecturesthat if e is a p-block of a finite group G with abelian defectgroup D and if f is the Brauer correspondent block of e of thenormalizer, N_G(D), of D then e and f have equivalent derivedcategories over a complete discrete valuation ring with residuefield of characteristic p. Some evidence for this conjecturehas been obtained using an important Morita analog for derivedcategories of Rickard [11]. This result states that the existenceof a tilting complex is a necessary and sufficient conditionfor the equivalence of two derived categories. In [5], Brouéalso defines an equivalence on the character level between p-blockse and f of finite groups G and H that he calls a ‘perfectisometry’ and he demonstrates that it is a consequenceof a derived category equivalence between e and f. In [5], Brouéalso poses a corresponding perfect isometry conjecture betweena p-block e of a finite group G with an abelian defect groupD and its Brauer correspondent p-block f of N_G(D) and presentsseveral examples of this phenomena. Subsequent research hasprovided much more evidence for this character-level conjecture. In many known examples of a perfect isometry between p-blockse, f of finite groups G, H there are also perfect isometriesbetween p-blocks of p-local subgroups corresponding to e andf and these isometries are compatible in a precise sense. In[5], Broué calls such a family of compatible perfectisometries an ‘isotypy’. In [11], Rickard addresses the analogous question of defininga p-locally compatible family of derived equivalences. In thisimportant paper, he defines a ‘splendid tilting complex’for p-blocks e and f of finite groups G and H with a commonp-subgroup P. Then he demonstrates that if X is such a splendidtilting complex, if P is a Sylow p-subgroup of G and H and ifG and H have the same ‘p-local structure’, thenp-local splendid tilting complexes are obtained from X via theBrauer functor and ‘lifting’. Consequently, in thissituation, we obtain an isotypy when e and f are the principalblocks of G and H. Linckelmann [9] and Puig [10] have also obtained important resultsin this area. In this paper, we refine the methods and program of [11] toobtain variants of some of the results of [11] that have widerapplicability. Indeed, suppose that the blocks e and f of Gand H have a common defect group D. Suppose also that X is asplendid tilting complex for e and f and that the p-local structureof (say) H with respect to D is contained in that of G, thenthe Brauer functor, lifting and ‘cutting’ by blockindempotents applied to X yield local block tilting complexesand consequently an isotypy on the character level. Since thep-local structure containment hypothesis is satisfied, for example,when H is a subgroup of G (as is the case in Broué'sconjectures) our results extend the applicability of these ideasand methods.     

On the Convergence of Sequences of Rational Approximations to Analytic Functions of a Certain Class

Complex analytic methods based on the theory of Walsh (1935)and on properties of orthonormal polynomials with general weight-functionon [–1, 1] are applied to the construction of variousrational approximations on the interval [–k, k] to a functiong(x) defined by or by where Remainder estimates are obtained, and from these, in the caseswhere g(x) is real on [–k, k], an asymptotic formula isobtained for the maximum error of the best rational approximationin the sense of the uniform norm. It is also shown that therate of convergence of the sequence of best approximations ofdegree n is twice the minimum rate predicted by Walsh's theory,in the sense that the degree of the approximation required fora given precision is approximately only half as great. Graphs are shown which illustrate that, for a simple example,the remainder estimates on which this asymptotic formula isbased are remarkably accurate even for approximations of lowdegree.     

Non-Regular Tangential Behaviour of a Monotone Measure

A Radon measure µ on Rⁿ is said to be k-monotone if is a non-decreasing function on (0,) for every x Rⁿ. (If µ is the k-dimensional Hausdorffmeasure restricted to a k-dimensional minimal surface then thisimportant property is expressed by the monotonicity formula.)We give an example of a 1-monotone measure µ in R² withnon-unique and non-conical tangent measures at a point. Furthermore,we show that µ can be the one-dimensional Hausdorff measurerestricted to a closed set A R². 2000 Mathematics Subject Classification49Q05, 49Q20 (primary), 28A75, 53A10 (secondary).     

On the Growth of Ideals and Submodules

All rings in this paper are commutative, with identity. If thering R is either finitely generated, or semi-local (by whichI mean Noetherian and possessing only finitely many maximalideals), then R has only a finite number of ideals of each finiteindex. Thus it makes sense to study the function na_n(R), wherea_n(R) denotes the number of ideals of index n in R. In an earliernote [10], I determined a_n(R) for certain 2-dimensional domainsR. The results to be discussed here are less precise, but moregeneral; in particular, they deal with submodules of a finitelygenerated module. This makes it possible to translate them intoresults about the growth of normal subgroups in certain metabeliangroups, the original motivation for this work: for example,we determine the finitely generated metabelian groups with ‘polynomialnormal subgroup growth’ (Theorem 6.1).