Bregman distances and Klee sets (Q1028402): Difference between revisions

From MaRDI portal
Set OpenAlex properties.
ReferenceBot (talk | contribs)
Changed an Item
 
(One intermediate revision by one other user not shown)
Property / arXiv ID
 
Property / arXiv ID: 0802.2322 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Sets with unique farthest points / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4347616 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Bregman distances and Chebyshev sets / rank
 
Normal rank
Property / cites work
 
Property / cites work: Convex analysis and nonlinear optimization. Theory and examples. / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5580053 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Totally convex functions for fixed points computation and infinite dimensional optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4718793 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4000274 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Qualitative properties of trajectories of control systems: a survey / rank
 
Normal rank
Property / cites work
 
Property / cites work: Best approximation in inner product spaces / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5295627 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Potpourri of Conjectures and Open Questions in Nonlinear Analysis and Optimization / rank
 
Normal rank
Property / cites work
 
Property / cites work: Convexity of Chebyshev sets / rank
 
Normal rank
Property / cites work
 
Property / cites work: On the multiplicity of Dini subgradients in separable spaces / rank
 
Normal rank
Property / cites work
 
Property / cites work: The number of farthest points / rank
 
Normal rank
Property / cites work
 
Property / cites work: On farthest points of sets / rank
 
Normal rank
Property / cites work
 
Property / cites work: Convex functions, monotone operators and differentiability. / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5638112 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Variational Analysis / rank
 
Normal rank
Property / cites work
 
Property / cites work: Minimax monotonicity / rank
 
Normal rank
Property / cites work
 
Property / cites work: Duality for nonconvex optimization and its applications / rank
 
Normal rank
Property / cites work
 
Property / cites work: Farthest points and monotone operators / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3149262 / rank
 
Normal rank

Latest revision as of 17:15, 1 July 2024

scientific article
Language Label Description Also known as
English
Bregman distances and Klee sets
scientific article

    Statements

    Bregman distances and Klee sets (English)
    0 references
    0 references
    0 references
    0 references
    0 references
    30 June 2009
    0 references
    The authors extend the known ``farthest-point-assertion'' which points out that each Klee set in an Euclidean space (i.e. each nonempty bounded closed set \(C\subset\mathbb{R}^n\) for which each point \(y\in\mathbb{R}^n\) has an unique farthest point \(x\) in \(C\)) is singleton. In the paper, the Euclidean distance is replaced by so-called ``Bregman distances'' which are induced by smooth strictly convex functions \(f:\mathbb{R}^n\to\mathbb{R}\) (with \(C\subset\text{int\,dom}\, f\)) according to \[ D_f(x, y)= \begin{cases} f(x)- f(y)-\langle\nabla f(y), x-y\rangle, &\text{if }x,y\in \operatorname{dom}f,\\ +\infty, &\text{else}.\end{cases} \] Clearly, if \(f(x)= {1\over 2}\| x\|^2\) then it is \(D_f(x, y)={1\over 2}\| x-y\|^2\). In general, however, \(D_f(\cdot,\cdot)\) is not a distance function by definition since it is neither symmetric nor does the triangle inequality hold. The authors show that also Klee sets with respect of Bregman distances are singleton. They provide two proofs: the first proof bases on special results from monotone operator theory, the second proof uses generalized subdifferentials from nonsmooth analysis.
    0 references
    farthest-point-conjecture
    0 references
    Bregman distance
    0 references
    Bregman projection
    0 references
    Legendre function
    0 references
    maximal monotone operator
    0 references
    subdifferential
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references