Generalised bargaining sets for cooperative games

Although theM ₁ -bargaining set for games with side payments is known to exist, it frequently contains payoffs which are highly inequitable. For this reason the more restrictedM ₂-bargaining set is of interest. SinceM ₂ is not known to exist in general, this paper introduces anM _*-bargaining set, contained inM ₁ and containingM ₂, and presents an existence theorem. For the class of symmetric, simple games with decreasing returns, theM ₂ -bargaining set is shown to exist, and a fairly severe restriction on payoffs satisfyingM ₂ -stability is obtained.     

Circuits asymptotically optimal in reliability in special bases

We consider realization of Boolean functions by the circuits composed of unreliable elements in a complete basis B ? B ₃ (with B ₃ the set of all Boolean functions of three variables x ₁, x ₂, and x ₃). We assume that, with probability ? ∈ (0, 1/2), all elements of a circuit are independently subjected to inverse failures at outputs. All bases are found in which it is possible to realize almost all Boolean functions by the circuits asymptotically optimal in reliability and functioning with unreliability 3? as ? → 0. It is proved that there are no other bases B ? B ₃ like these.     

On Supersets of Wavelet Sets

Considering a single dyadic orthonormal wavelet ψ in L ²(?), it is still an open problem whether the support of $\widehat{\psi}$ always contains a wavelet set. As far as we know, the only result in this direction is that if the Fourier support of a wavelet function is “small” then it is either a wavelet set or a union of two wavelet sets. Without assuming that a set S is the Fourier support of a wavelet, we obtain some necessary conditions and some sufficient conditions for a “small” set S to contain a wavelet set. The main results, which are in terms of the relationship between two explicitly constructed subsets A and B of S and two subsets T ₂ and D ₂ of S intersecting itself exactly twice translationally and dilationally respectively, are (1) if $A\cup B\not\subseteq T_{2}\cap D_{2}$ then S does not contain a wavelet set; and (2) if A∪B?T ₂∩D ₂ then every wavelet subset of S must be in S?(A∪B) and if S?(A∪B) satisfies a “weak” condition then there exists a wavelet subset of S?(A∪B). In particular, if the set S?(A∪B) is of the right size then it must be a wavelet set.     

Representations of small braid groups in Temperley-Lieb algebra

In this paper we consider the question of faithfulness of the Jones' representation of braid group Bn into the Temperley-Lieb algebra TLn. The obvious motivation to study this problem is that any non-trivial element in the kernel of this representation (for any n) would almost certainly yield a non-trivial knot with trivial Jones polynomial (see [S. Bigelow, Does the Jones polynomial detect the unknot? J. Knot Theory Ramifications 11 (4) (2002) 493-505], we will explain it in more detail in Section 1). As one of the two main results we prove Theorem 1 in which we present a method to obtain non-trivial elements in the kernel of the representation of B₆ into TL9,2—to the authors' knowledge the first such examples in the second gradation of the Temperley-Lieb algebra. Theorem 2 which is a refinement of Theorem 1 may be used to produce smaller examples of the same kind. We also show briefly how some braids that are used in Section 4 to construct specific examples were generated with a computer program.     

Multiplicative spectrum of ultrametric Banach algebras of continuous functions

Let K be an ultrametric complete field and let E be an ultrametric space. Let A be the Banach K-algebra of bounded continuous functions from E to K and let B be the Banach K-algebra of bounded uniformly continuous functions from E to K. Maximal ideals and continuous multiplicative semi-norms on A (resp. on B) are studied by defining relations of stickiness and contiguousness on ultrafilters that are equivalence relations. So, the maximal spectrum of A (resp. of B) is in bijection with the set of equivalence classes with respect to stickiness (resp. to contiguousness). Every prime ideal of A or B is included in a unique maximal ideal and every prime closed ideal of A (resp. of B) is a maximal ideal, hence every continuous multiplicative semi-norms on A (resp. on B) has a kernel that is a maximal ideal. If K is locally compact, every maximal ideal of A (resp. of B) is of codimension 1. Every maximal ideal of A or B is the kernel of a unique continuous multiplicative semi-norm and every continuous multiplicative semi-norm is defined as the limit along an ultrafilter on E. Consequently, on A as on B the set of continuous multiplicative semi-norms defined by points of E is dense in the whole set of all continuous multiplicative semi-norms. Ultrafilters show bijections between the set of continuous multiplicative semi-norms of A, Max(A) and the Banaschewski compactification of E which is homeomorphic to the topological space of continuous multiplicative semi-norms. The Shilov boundary of A (resp. B) is equal to the whole set of continuous multiplicative semi-norms.     

The annihilator of Bergman space

LetD be a bounded plane domain (with some smoothness requirements on its boundary). LetB _p(D), 1≤p<∞, be the Bergmanp-space ofD. In a previous paper we showed that the “natural projection”P, involving the Bergman kernel forD, is a bounded projection fromL _p(D) ontoB _p(D), 1<p<∞. With this we have the decompositionL _p(D)=B _p(D)⊕B _q ^⊥ (D,p ^–1+q ^–=1, 1<p< ∞. Here, we show that the annihilatorB _q ^⊥ (D) is the space of allL _p-complex derivatives of functions belonging to Sobolev space and which vanish on the boundary ofD. This extends a result of Schiffer for the casep=2. We also study certain operators onL _p(D). Especially, we show that , whereI is the identity operator and ? is an operator involving the adjoint of the Bergman kernel. Other relationships relevant toB _q ^⊥ (D) are studied.     

A direct bijection for the Harer-Zagier formula

We give a combinatorial proof of Harer and Zagier's formula for the disjoint cycle distribution of a long cycle multiplied by an involution with no fixed points, in the symmetric group on a set of even cardinality. The main result of this paper is a direct bijection of a set B_p,k, the enumeration of which is equivalent to the Harer-Zagier formula. The elements of B_p,k are of the form (μ,π), where μ is a pairing on {1,…,2p}, π is a partition into k blocks of the same set, and a certain relation holds between μ and π. (The set partitions π that can appear in B_p,k are called “shift-symmetric”, for reasons that are explained in the paper.) The direct bijection for B_p,k identifies it with a set of objects of the form (ρ,t), where ρ is a pairing on a 2(p-k+1)-subset of {1,…,2p} (a “partial pairing”), and t is an ordered tree with k vertices. If we specialize to the extreme case when p=k-1, then ρ is empty, and our bijection reduces to a well-known tree bijection.     

On the risk of estimates for block decreasing densities

A density f=f(x₁,…,x_d) on [0,∞)^d is block decreasing if for each j∈{1,…,d}, it is a decreasing function of x_j, when all other components are held fixed. Let us consider the class of all block decreasing densities on [0,1]^d bounded by B. We shall study the minimax risk over this class using n i.i.d. observations, the loss being measured by the L₁ distance between the estimate and the true density. We prove that if S=log(1+B), lower bounds for the risk are of the form C(S^d/n)^1/(d+2), where C is a function of d only. We also prove that a suitable histogram with unequal bin widths as well as a variable kernel estimate achieve the optimal multivariate rate. We present a procedure for choosing all parameters in the kernel estimate automatically without loosing the minimax optimality, even if B and the support of f are unknown.     

Berezin kernels and analysis on Makarevich spaces

Following ideas of van Dijk and Hille we study the link which exists between maximal degenerate representations and Berezin kernels.We consider the conformal group Conf(V) of a simple real Jordan algebra V. The maximal degenerate representations π_s (s ε ℂ) we shall study are induced by a character of a maximal parabolic subgroup P¯ of Conf(V). These representations π_s can be realized on a space I_s of smooth functions on V. There is an invariant bilinear form ℬ_s on the space Is. The problem we consider is to diagonalize this bilinear form ℬ_s, with respect to the action of a symmetric subgroup G of the conformal group Conf(V). This bilinear form can be written as an integral involving the Berezin kernel B_v an invariant kernel on the Riemannian symmetric space G/K, which is a Makarevich symmetric space in the sense of Bertram. Then we can use results by van Dijk and Pevzner who computed the spherical Fourier transform of B_v. From these, one deduces that the Berezin kernel satisfies a remarkable Bernstein identity: D(_ν)B_ν=b(ν)B_ν+1, where D(_ν) is an invariant differential operator on G/K and b(ν) is a polynomial. By using this identity we compute a Hua type integral which gives the normalizing factor for an intertwining operator from I_−s to I_s. Furthermore, we obtain the diagonalization of the invariant bilinear form with respect to the action of the maximal compact group U of the conformal group Conf(V).     

On bases with unreliability coefficient 2

Consider the realization of Boolean functions by networks from unreliable functional components in a complete basis B ? B ₃ (B ₃ is the set of all Boolean functions depending on the variables x ₁, x ₂, x ₃). It is assumed that all the components of the network are subject to inverse faults at the outputs independently of each other with probability ? ∈ (0, 1/2). In B ₃, we obtain all complete bases in which the following two conditions simultaneously hold: 1) any function can be realized by a network with unreliability asymptotically not greater than 2? (? → 0); 2) there exist functions (denote their set by K) that cannot be realized by networks with unreliability asymptotically less than 2?, ? → 0. Such bases will be called bases with unreliability coefficient 2. It is also proved that the set K contains almost all functions.     

Near-Boundary Asymptotics for Correlation Kernels

We discuss asymptotic formulas for the correlation kernel corresponding to a (more or less) general potential Q in the plane. If K _n is such a kernel, there is a known asymptotic formula K _n (z,z)=nΔQ(z)+B ₁(z)+n ^?1 B ₂(z)+? valid for z in the interior of a certain compact set known as the “droplet” corresponding to Q (on which ΔQ>0). On the other hand, if z is in the exterior of the droplet, K _n (z,z) converges to zero exponentially in n. Results of this type are useful in random matrix theory and conformal field theory; they have recently been used to prove the Gaussian field convergence in the interior of the droplet for fluctuations of eigenvalues of random normal matrices. To be able to extend such results beyond the interior, it becomes necessary to have a certain uniformity of estimates as z approaches the boundary either from the interior or from the exterior. Such estimates have to our knowledge hitherto not been known on a rigorous level. This note intends to fill this gap. We will consider applications in later publications. Our treatment of the (interior) asymptotics relies in an essential way on previous work due to Berman, Berndtsson, and Sjöstrand (Ark. Mat. 46, 2008), and Berman (Indiana Univ. Math. J. 58, 2009). We hope that this note can to some extent be regarded as a contribution to that work.     

A relationship between stochastic matrices and eigenvalues of products

Let {B₁,…,B_n} be a set of n rank one n×n row stochastic matrices. The next two statements are equivalent: (1) If A is an n×n nonnegative matrix, then 1 is an eigenvalue ofB_kA for each k=1,…,n if and only if A is row stochastic. (2) The n×n row stochastic matrix S whose kth row is a row of B_k for k=1,…,n is nonsingular. For any set {B₁, B₂,…, B_p} of fewer than n row stochastic matrices of order n×n and of any rank, there exists a nonnegative n×n matrix A which is not row stochastic such that 1 is eigenvalue of every B_kA, k=1,…,p.     

The decomposition of Cr(K) into the direct sum of subalgebras

Let A and B be closed subalgebras of C_r(X) whose direct sum is C_r(X). Some consequences of this relation are explored in this paper. For example if 1 ?A (as may be assumed) it is shown that the norm of the projection onto A is an odd integer and there is a retraction of X onto the set of common zeros of elements of B.     

The harmonic Bergman kernels for punctured domains

We obtain the explicit formulae for the harmonic Bergman kernels of Bn／{0} and Rn／Bn and study the connection between harmonic Bergman kernel and weighted harmonic Bergman kernel.We also get the explicit formula for the weighted harmonic Bergman kernel of Bn／{0} with the weight 1/｜x｜4.     

The group of parenthesized braids

We investigate a group B_• that includes Artin's braid group B_∞ and Thompson's group F. The elements of B_• are represented by braids diagrams in which the distances between the strands are not uniform and, besides the usual crossing generators, new rescaling operators shrink or stretch the distances between the strands. We prove that B_• is a group of fractions, that it is orderable, admits a nontrivial self-distributive structure, i.e., one involving the law x(yz)=(xy)(xz), it embeds in the mapping class group of a sphere with a Cantor set of punctures, and that Artin's representation of B_∞ into the automorphisms of a free group extends to B_•.     

Balanced bipartite weighing designs

A block B denotes a set of k = k₁ + k₂ elements which are divided into two subsets B¹ and B², where |Bⁱ| = k_i, i = 1 or 2. Two elements of B are said to be linked or n-linked in B if and only if they belong to different subsets or the same subset of B respectively. A balanced bipartite weighing design, (briefly BBWD (υ, k₁, k₂, λ₁)) is an arrangement of υ elements into b blocks, each containing k elements, such that each element occurs in exactly r blocks, any two distinct elements are linked in exactly λ₁ blocks and n-linked in exactly λ₂ blocks.Given fixed k₁ and k₂, there is always a minimal value of λ₁ such that the necessary conditions for the existence of a BBWD are satisfied for same υ. It is proved that in many cases, the necessary conditions are also sufficient. Some general methods for constructing BBWD's as well as a table of all designs with υ ? 13 are obtained.     

Maximizing the probability of correctly ordering random variables using linear predictors

Let (T₁, x₁), (T₂, x₂), …, (T_n, x_n) be a sample from a multivariate normal distribution where T_i are (unobservable) random variables and x_i are random vectors in R^k. If the sample is either independent and identically distributed or satisfies a multivariate components of variance model, then the probability of correctly ordering {T_i} is maximized by ranking according to the order of the best linear predictors {E(T_i|x_i)}. Furthermore, it orderings are chosen according to linear functions {b′x_i} then the conditional probability of correct order given (T_i = t₁; i = 1, …, n) is maximized when b′x_i is the best linear predictor. Examples are given to show that linear predictors may not be optimal and that using a linear combination other that the best linear predictor may give a greater probability of correctly ordering {T_i} if {(T_i, x_i)} are independent but not identically distributed, or if the distributions are not normal.     

Hall-type representations for generalised orthogroups

An orthogroup is a completely regular orthodox semigroup. The main purpose of this paper is to find a representation of a (generalised) orthogroup with band of idempotents B in terms of a fundamental (generalised) orthogroup. The latter is a subsemigroup of the Hall semigroup W _B (or of its generalisations V _B ,U _B and S _B ). We proceed in the regular case by constructing a fundamental completely regular subsemigroup \(\overline{W_{B}}\) of W _B , using two different methods. Our subsemigroup plays the role for orthogroups that W _B plays for orthodox semigroups, in that it contains a representation of every orthogroup with band of idempotents B, with kernel of the representation being μ, the greatest congruence contained in \(\mathcal{H}\) . To develop an analogous theory for classes of generalised orthogroups, that is, to extend beyond the regular case, we replace \(\mathcal{H}\) by \(\widetilde{\mathcal{H}}_{B}\) . Generalised orthogroups are then classes of weakly B-superabundant semigroups with (C). We first consider those satisfying an idempotent connected condition (IC) or (WIC). We construct fundamental weakly B-superabundant subsemigroups \(\overline{V_{B}}\) (respectively, \(\overline{U_{B}}\) ) of V _B (respectively, U _B ) with (C) and (IC) (respectively, with (C) and (WIC)) such that any weakly B-superabundant semigroup with (C) and (IC) (respectively, with (C) and (WIC)) admits a representation to \(\overline{V_{B}}\) (respectively, \(\overline{U_{B}}\) ), with kernel of the respresentation being μ _B , the greatest congruence contained in \(\widetilde{\mathcal{H}}_{B}\) . Finally, we remove the idempotent connected condition and find a representation for an arbitrary weakly B-superabundant semigroup with (C), making use of fresh technology, constructing a fundamental weakly B-superabundant subsemigroup \(\overline{S_{B}}\) of S _B , with the appropriate universal properties. We note that our results are needed in a parallel paper to complete the representation of arbitrary weakly B-superabundant semigroups with (C) as spined products of superabundant Ehresmann semigroups and subsemigroups of S _B .     

Average boundary conditions in Cauchy problems

We consider the general Cauchy problem with initial data in a Hilbert space and with a formal dissipative linear generator. A complete parametrization is known of the (abstract) boundary conditions which make this problem well set. We exhibit a distinguished subset $\text{B}$_E of the set $\text{B}$ of boundary conditions and demonstrate explicitly that the evolution associated with each B in $\text{B}$ can be represented as a (time independent) average over the evolutions associated with B′ in $\text{B}$_E. Applications are discussed to Schrödinger equations in bounded regions or with singular potentials.