Scribd Logo
Upload

Welcome to Scribd!

  • 20 views

    Convexity of Ceby Sev Sets in Hilbert Spaces: Int. J. Contemp. Math. Sciences, Vol. 4, 2009, No. 12, 573 - 576

    Uploaded by

    noiaybinhyen277
    This document summarizes research on the convexity of Chebyshev sets in Hilbert spaces. It begins by introducing Chebyshev sets and noting several existing results about their convexity in finite-dimensional and smooth normed spaces. The main results proved are that every weakly closed Chebyshev set in a uniformly convex Banach space is convex, and therefore every weakly closed Chebyshev set in a Hilbert space is convex. This resolves an open problem about whether all Chebyshev sets in Hilbert spaces must be convex. The proofs rely on properties of Kadec norms, strict convexity, smoothness, and continuity of the metric projection.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    Convexity of Ceby Sev Sets in Hilbert Spaces: Int. J. Contemp. Math. Sciences, Vol. 4, 2009, No. 12, 573 - 576

    Uploaded by

    noiaybinhyen277
    20 views4 pages
    This document summarizes research on the convexity of Chebyshev sets in Hilbert spaces. It begins by introducing Chebyshev sets and noting several existing results about their convexity in finite-dimensional and smooth normed spaces. The main results proved are that every weakly closed Chebyshev set in a uniformly convex Banach space is convex, and therefore every weakly closed Chebyshev set in a Hilbert space is convex. This resolves an open problem about whether all Chebyshev sets in Hilbert spaces must be convex. The proofs rely on properties of Kadec norms, strict convexity, smoothness, and continuity of the metric projection.

    Original Title

    461AA452d01

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document summarizes research on the convexity of Chebyshev sets in Hilbert spaces. It begins by introducing Chebyshev sets and noting several existing results about their convexity in finite-dimensional and smooth normed spaces. The main results proved are that every weakly closed Chebyshev set in a uniformly convex Banach space is convex, and therefore every weakly closed Chebyshev set in a Hilbert space is convex. This resolves an open problem about whether all Chebyshev sets in Hilbert spaces must be convex. The proofs rely on properties of Kadec norms, strict convexity, smoothness, and continuity of the metric projection.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    20 views4 pages

    Convexity of Ceby Sev Sets in Hilbert Spaces: Int. J. Contemp. Math. Sciences, Vol. 4, 2009, No. 12, 573 - 576

    Uploaded by

    noiaybinhyen277
    This document summarizes research on the convexity of Chebyshev sets in Hilbert spaces. It begins by introducing Chebyshev sets and noting several existing results about their convexity in finite-dimensional and smooth normed spaces. The main results proved are that every weakly closed Chebyshev set in a uniformly convex Banach space is convex, and therefore every weakly closed Chebyshev set in a Hilbert space is convex. This resolves an open problem about whether all Chebyshev sets in Hilbert spaces must be convex. The proofs rely on properties of Kadec norms, strict convexity, smoothness, and continuity of the metric projection.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 4

    Int. J. Contemp. Math. Sciences, Vol. 4, 2009, no.

    12, 573 - 576

    Convexity of C̆ebys̆ev Sets


    in Hilbert Spaces
    H. Haghshenas and M. R. Farmani

    Department of Mathematics, Birjand University, Iran


    h− [email protected], [email protected]

    Abstract. The aim of this paper is state of conditions that ensure the
    convexity of a C̆ebys̆ev sets in Hilbert spaces.

    Mathematics Subject Classification: 46B20

    Keywords: distance function, C̆ebys̆ev set, metric projection, Kadec norm,


    smooth space, strictly convex space, uniformly convex space

    1. Introduction
    The approximation theory is one of the important branch of functional
    analysis that C̆ebys̆ev originated it in nineteenth century. But, convexity of
    C̆ebys̆ev sets is one of the basic problems in this thoery. In a finite dimensional
    smooth normed linear space a C̆ebys̆ev set is convex[2]. Also, every boundedly
    compact C̆ebys̆ev set in a smooth Banach space is convex [3,7] and in a Banach
    space which is uniformly smooth, each approximately compact C̆ebys̆ev set is
    convex[4]. In addition, in a strongly smooth space, every C̆ebys̆ev set with
    continuous metric projection is convex[5,6]. Regarding convexity of C̆ebys̆ev
    sets, there are still several open problems. It is a wellknown problem whether
    a C̆ebys̆ev set in a Hilbert space must be convex. Of course, in a finite-
    dimensional Hilbert space, every C̆ebys̆ev set is convex. For inverse, we know
    that every closed convex set in a strictly convex reflexive Banach space and
    in particular Hilbert space is C̆ebys̆ev. However, this problem that whether
    every C̆ebys̆ev set in a strictly convex reflexive Banach space is convex is still
    open.
    574 H. Haghshenas and M. R. Farmani
    2. Basic definitions and Preliminaries
    In this section we collect some elementary facts which will help us to establish
    our main results.
    Definition 2.1. Let (X, .) be a real normed linear space, x ∈ X and X ∗
    be its dual space. For a nonempty subset K in X, the distance of x from K
    is defined as dK (x) = inf {x − v; v ∈ K}. The K is said to be a C̆ebys̆ev
    set if, each point in X has a unique best approximation in K. In other words,
    for every x ∈ X, there exist a unique v ∈ K such that x − v = dK (x).
    (This concept was introduced by S. B. Stechkin in honour of the founder
    of best approximation theory, C̆ebys̆ev). The metric projection is given by
    PK (x) = {v ∈ K; x − v = dK (x)} which consists of the closest points in
    K to x. The PK is said to be continuous if, PK (x) is a singleton for each
    x ∈ X \ K and it is sequently continuous.
    Definition 2.2. A norm . on X is said to be Kadec if, each weakly
    convergent sequence (xn )∞ n=1 in X with the weak limit x ∈ X converges in
    norm to x whenever xn  → x as n → ∞.
    Definition 2.3. The space X is said to be strictly convex if, x = y whenever
    x+y
    x, y ∈ S(X) and ∈ S(X), where S(X) = {x ∈ X; x = 1}.
    2
    Remark 2.4. Every Hilbert space is strictly convex. Hence the dual of each
    Hilbert space is strictly convex.
    Related to the notion of strict convexity, is the notion of smoothness.
    Definition 2.5. For each x ∈ X the element x∗ ∈ S(X ∗ ) satisfying x =
    x∗ , x is called the support functional corresponding to x and X is smooth in
    a non-zero x ∈ X if, the support functional corresponding to x is unique.
    Of course, the Hahn-Banach extension theorem, ensures the existence of at
    least one such support functional.
    Smoothness and strict convexity are not quite dual properties. There are
    examples of smooth spaces whose duals fail to be strictly convex.
    Theorem 2.6. [1] Each Hilbert space is smooth.
    Example 2.7. The space R2 with Euclidian norm,
    
    (x1 , x2 ) = x21 + x22
    is a smooth and strictly convex space.

    Theorem 2.8. [1] If X be a reflexive and smooth space, then the dual space
    X ∗ is strictly convex.
    Definition 2.9. The space X is uniformly convex if, for every sequences
    (xn )∞ ∞
    n=1 and (yn )n=1 we have, lim xn − yn  = 0 whenever, lim xn + yn  = 2
    n→∞ n→∞
    Theorem 2.10. [1] Every uniformly convex space, is strictly convex.
    Convexity of C̆ebys̆ev sets in Hilbert spaces 575
    Remark 2.11. [1] The inverse of theorem (2.10) is not true, necessary. For
    example, define a norm |.| on C[0, 1] by |x|2 = x2∞ + x22 , where .∞
    and .2 , denote the norms of C[0, 1] and L2 [0, 1], respectively. Then |.| is
    strictly convex but, not uniformly convex on C[0, 1].
    Theorem 2.12. [1] Each Hilbert space, is uniformly convex.
    Theorem 2.13. [1] Every uniformly convex Banach space, is reflexive.
    Theorem 2.14. [1] The norm of every uniformly convex space, is Kadec.

    3. Main Results
    In this section, we state the conditions that ensure the convexity of a C̆ebys̆ev
    sets in Hilbert spaces
    Theorem 3.1. Let K be a weakly closed set in a reflexive space X with
    Kadec norm. Then the metric projection PK is continuous.
    Proof. Let x ∈ X\K, v ∈ PK (x) and suppose (xn )∞ ∞
    n=1 ⊆ X, (vn )n=1 ⊆ PK (xn )
    such that xn → x in norm. It is sufficient show that PK (x) is a singleton and
    vn → v in norm. Since that dK is continuous, we have :
    dK (x) = x−v ≤ vn −x ≤ xn −vn +xn −x = dK (xn )+xn −x → dK (x)
    So, lim vn − x = x − v and hence (vn )∞ ∞
    n=1 is bounded. Thus (vn )n=1 is
    n→∞
    contained in an set A such that A is weakly closed and boundedly in norm.
    Since X is reflexive, the set A is weakly compact. Hence there exists a weakly
    convergent subsequence (vnk )∞ ∞
    k=1 of (vn )n=1 whose weak limit v0 lies in A:
    Such an v0 must be in K. Note that the norm on a normed space is lower
    semicontinuous for the weak topology. Then
    x − v = dK (x) ≤ x − v0  ≤ lim inf vnk − x = dk (x) = x − v
    k→∞

    This implies v0 = v and so PK (x) is a singleton. The (x − vnk )∞


    k=1 is weakly
    converges to x−v and satisfies lim vnk − x = x − v. Since the norm on X
    k→∞
    is Kadec, the sequence (x − vnk )∞
    k=1 is normly convergent to (x − v). Therefore,
    (vnk )k=1 converges to v in norm and consequently (vn )∞
    ∞
    n=1 converges to v in
    norm . This proves that PK is continuous.
    Now by the theorems (2.12), (2.13), (3.1), we have:
    Corollary 3.2. Let K be a weakly closed set in a uniformly convex Banach
    space X. Then the metric projection PK is continuous.
    Theorem 3.3. [5,6] Every C̆ebys̆ev set K with continuous metric projection
    PK , in a Banach space X with strictly convex dual X ∗ , is convex.
    Now by remark (2.4) and the previous theorem, we have:
    Corollary 3.4. Every C̆ebys̆ev set with continuous metric projection, in a
    Hilbert space is convex.
    576 H. Haghshenas and M. R. Farmani
    Now by the theorems (3.3), (2.12), (2.8) and corollary (3.2), we have:
    Theorem 3.5. Every weakly closed C̆ebys̆ev set in a smooth uniformly
    convex Banach space, is convex.
    Finally, by the theorems (2.6), (2.11) and the previous theorem, we have:
    Corollary 3.6. Every weakly closed C̆ebys̆ev set in a Hilbert space, is convex
    .

    4. Acknowledgments
    We wish to express our appreciation to our supervisor Dr. A. Assadi for his
    advice and encouragement in the preparation of this dissertation.

    References
    [1] L. Carothers, Short course on Banach space theory, Cambridge University Press, 2005.
    [2] V. Klee, Convexity of Chebyshev sets, Math. Annal. 142 (1961), 292-304.
    [3] L. P. Vlasov, Chebyshev sets in Banach spaces, Soviet Math. Dokl. 2(1961), 1373-1374.
    [4] L. P. Vlasov, Approximately convex sets in uniformly smooth spaces, Math. Notes
    Acad. Sci. USSR 1(1967), 295-299.
    [5] L. P. Vlasov, Chebyshev sets and approximately convex sets, Math. Notes Acad. Sci.
    USSR 2(1967), 600-605.
    [6] L. P. Vlasov, Almost convex and Chebyshev sets, Math. Notes Acad. Sci. USSR 8(1970),
    776-779.
    [7] L. P. Vlasov, Approximative properties of sets in normed linear spaces, Russian Math.
    Surveys 28(1973), 1-66.

    Received: September, 2008

    You might also like