Currently displaying 1 – 19 of 19

Showing per page

Order by Relevance | Title | Year of publication

Bounds for f -divergences under likelihood ratio constraints

Sever Silvestru Dragomir — 2003

Applications of Mathematics

In this paper we establish an upper and a lower bound for the f -divergence of two discrete random variables under likelihood ratio constraints in terms of the Kullback-Leibler distance. Some particular cases for Hellinger and triangular discimination, χ 2 -distance and Rényi’s divergences, etc. are also considered.

Approximation of the Stieltjes integral and applications in numerical integration

Pietro CeroneSever Silvestru Dragomir — 2006

Applications of Mathematics

Some inequalities for the Stieltjes integral and applications in numerical integration are given. The Stieltjes integral is approximated by the product of the divided difference of the integrator and the Lebesgue integral of the integrand. Bounds on the approximation error are provided. Applications to the Fourier Sine and Cosine transforms on finite intervals are mentioned as well.

Two mappings related to semi-inner products and their applications in geometry of normed linear spaces

Sever Silvestru DragomirJaromír J. Koliha — 2000

Applications of Mathematics

In this paper we introduce two mappings associated with the lower and upper semi-inner product ( · , · ) i and ( · , · ) s and with semi-inner products [ · , · ] (in the sense of Lumer) which generate the norm of a real normed linear space, and study properties of monotonicity and boundedness of these mappings. We give a refinement of the Schwarz inequality, applications to the Birkhoff orthogonality, to smoothness of normed linear spaces as well as to the characterization of best approximants.

Norm inequalities for the difference between weighted and integral means of operator differentiable functions

Silvestru Sever Dragomir — 2020

Archivum Mathematicum

Let f be a continuous function on I and A , B 𝒮𝒜 I H , the convex set of selfadjoint operators with spectra in I . If A B and f , as an operator function, is Gateaux differentiable on [ A , B ] : = ( 1 - t ) A + t B t 0 , 1 , while p : 0 , 1 is Lebesgue integrable, then we have the inequalities 0 1 p τ f 1 - τ A + τ B d τ - 0 1 p τ d τ 0 1 f 1 - τ A + τ B d τ 0 1 τ ( 1 - τ ) | τ 1 p s d s 1 - τ - 0 τ p s d s τ | f 1 - τ A + τ B B - A d τ 1 4 0 1 | τ 1 p s d s 1 - τ - 0 τ p s d s τ | f 1 - τ A + τ B B - A d τ , where f is the Gateaux derivative of f .

Bound on extended f -divergences for a variety of classes

The concept of f -divergences was introduced by Csiszár in 1963 as measures of the ‘hardness’ of a testing problem depending on a convex real valued function f on the interval [ 0 , ) . The choice of this parameter f can be adjusted so as to match the needs for specific applications. The definition and some of the most basic properties of f -divergences are given and the class of χ α -divergences is presented. Ostrowski’s inequality and a Trapezoid inequality are utilized in order to prove bounds for an extension...

Page 1

Download Results (CSV)