A Bang-Bang representation for 3×3 embeddable stochastic matrices

Summary It is proved that a 3×3 embeddable stochastic matrix has a representation as a product of a finite number of elementary stochastic matrices, with only one off-diagonal element positive. In particular if the determinant is 1/2 then only 6 matrices are needed and a necessary and sufficient condition for embeddability in this case is given.     

A geometrical method for diagonalizing real,symmetric 3×3 matrices through Euler rotations

The classical problem of finding the principal values (eigenvalues) and principal axes (eigenvectors) of a physical property represented by a second-rank symmetric tensor is treated in textbooks by solving the characteristic equation associated with the 3×3 symmetric matrix representation. The same problem is solved here without reference to the characteristic equation. By use of Euler rotations, analytical expressions are attained for the Euler eigenangles, the eigenvalues and the eigenvectors.     

A new method for computing Moore–Penrose inverse matrices

The Moore–Penrose inverse of an arbitrary matrix (including singular and rectangular) has many applications in statistics, prediction theory, control system analysis, curve fitting and numerical analysis. In this paper, an algorithm based on the conjugate Gram–Schmidt process and the Moore–Penrose inverse of partitioned matrices is proposed for computing the pseudoinverse of an m×n$m\times n$ real matrix A$A$ with m≥n$m\ge n$ and rank r≤n$r\le n$. Numerical experiments show that the resulting pseudoinverse matrix is reasonably accurate and its computation time is significantly less than that of pseudoinverses obtained by the other methods for large sparse matrices.     

Handelman’s Positivstellensatz for polynomial matrices positive definite on polyhedra

In this paper we give a matrix version of Handelman’s Positivstellensatz (Handelman in Pac J Math 132:35–62, 1988), representing polynomial matrices which are positive definite on convex, compact polyhedra. Moreover, we propose also a procedure to find such a representation. As a corollary of Handelman’s theorem, we give a special case of Schmüdgen’s Positivstellensatz for polynomial matrices positive definite on convex, compact polyhedra.     

Indefinite numerical range of 3 × 3 matrices

The point equation of the associated curve of the indefinite numerical range is derived, following Fiedler’s approach for definite inner product spaces. The classification of the associated curve is presented in the 3 × 3 indefinite case, using Newton’s classification of cubic curves. Illustrative examples of all the different possibilities are given. The results obtained extend to Krein spaces results of Kippenhahn on the classical numerical range. The work of this author was partially supported by the Portuguese foundation FCT, in the scope of program POCI 2010.     

Lorentzian geometry for 2 × 2 real matrices

We give an interpretation and a natural proof of the lemma of Herglotz on 2 × 2 symmetric matrices from the point of view of 3-dimensional Lorentz geometry and 2-dimensional hyperbolic geometry. This interpretation leads naturally to an analogous result on 2 × 2 matrices with trace 0.     

Injective comodules for 2 × 2 quantum matrices

We study the injective comodules for the coordinate bialgebra of quantum 2 × 2 matrices at a root of unity. The related GL ₂ and SL ₂ theories are also specified.     

On a minimal set of generators for the invariants of 3 × 3 matrices

Let K be a field of characteristic p > 0, let L be a restricted Lie algebra and let R be an associative K-algebra. It is shown that the various constructions in the literature of crossed product of R with u(L) are equivalent. We calculate explicit formulae relating the parameters involved and obtain a formula which hints at a noncommutative version of the Bell polynomials.     

A new class of polynomial primal–dual methods for linear and semidefinite optimization

We propose a new class of primal–dual methods for linear optimization (LO). By using some new analysis tools, we prove that the large-update method for LO based on the new search direction has a polynomial complexity of O(n^4/(4+ρ)log(n/ε)) iterations, where ρ∈[0,2] is a parameter used in the system defining the search direction. If ρ=0, our results reproduce the well-known complexity of the standard primal–dual Newton method for LO. At each iteration, our algorithm needs only to solve a linear equation system. An extension of the algorithms to semidefinite optimization is also presented.     

Criteria for unitary congruence of complex 2 × 2 matrices

Canonical forms of 2 × 2 matrices with respect to unitary congruence transformations are described. This makes it possible to formulate simple criteria for checking whether the given matrices are unitarily congruent.     

Some generators for the algebra of n×n matrices

In this note, we show how the algebra of n×n matrices over a field can be generated by a pair of matrices A B, where A is an arbitrary nonscalar matrix and B can be chosen so that there is the maximum degree of linear independence between the higher commutators of B with A.     

A new non‐polynomial univariate interpolation formula of Hermite type

A new C ^∞ interpolant is presented for the univariate Hermite interpolation problem. It differs from the classical solution in that the interpolant is of non‐polynomial nature. Its basis functions are a set of simple, compact support, transcendental functions. The interpolant can be regarded as a truncated Multipoint Taylor series. It has essential singularities at the sample points, but is well behaved over the real axis and satisfies the given functional data. The interpolant converges to the underlying real‐analytic function when (i) the number of derivatives at each point tends to infinity and the number of sample points remains finite, and when (ii) the spacing between sample points tends to zero and the number of specified derivatives at each sample point remains finite. A comparison is made between the numerical results achieved with the new method and those obtained with polynomial Hermite interpolation. In contrast with the classical polynomial solution, the new interpolant does not suffer from any ill conditioning, so it is always numerically stable. In addition, it is a much more computationally efficient method than the polynomial approach. This revised version was published online in June 2006 with corrections to the Cover Date.     

Group inverses for some 2 × 2 block matrices over rings

We first consider the group inverses of the block matrices (A0BC) over a weakly finite ring. Then we study the sufficient and necessary conditions for the existence and the representations of the group inverses of the block matrices (ACBD) over a ring with unity 1 under the following conditions respectively: (i) B = C, D = 0, B^# and (B^πA) ^# both exist; (ii) B is invertible and m = n; (iii) A^# and (D - CA^#B)^# both exist, C = CAA^# , where A and D are m × m and n × n matrices, respectively.     

Jacobi method for symmetric 4 × 4 matrices converges for every cyclic pivot strategy

The paper studies the global convergence of the Jacobi method for symmetric matrices of size 4. We prove global convergence for all 720 cyclic pivot strategies. Precisely, we show that inequality S(A ^[t+3]) ≤ γ S(A ^[t]), t ≥ 1, holds with the constant γ < 1 that depends neither on the matrix A nor on the pivot strategy. Here, A ^[t] stands for the matrix obtained from A after t full cycles of the Jacobi method and S(A) is the off-diagonal norm of A. We show why three consecutive cycles have to be considered. The result has a direct application on the J-Jacobi method.     

A Frank–Wolfe type theorem for nondegenerate polynomial programs

In this paper, we study the existence of optimal solutions to a constrained polynomial optimization problem. More precisely, let \(f_0\) and \(f_1, \ldots , f_p :{\mathbb {R}}^n \rightarrow {\mathbb {R}}\) be convenient polynomial functions, and let \(S := \{x \in {\mathbb {R}}^n \ : \ f_i(x) \le 0, i = 1, \ldots , p\} \ne \emptyset .\) Under the assumption that the map \((f_0, f_{1}, \ldots , f_{p}) :{\mathbb {R}}^n \rightarrow {\mathbb {R}}^{p + 1}\) is non-degenerate at infinity, we show that if \(f_0\) is bounded from below on \(S,\) then \(f_0\) attains its infimum on \(S.\)