Posts by Costica Mustata

Costica Mustata

Abstract

Authors

Costica Mustata
“Tiberiu Popoviciu” Institute of Numerical Analysis, Romanian Academy, Romania

Keywords

Paper coordinates

C. Mustăţa, On the extension of Lipschitz functions, Seminar of Funct. Analysis and Numerical Methods, Preprint Nr. 1 (1986), 83-92.

PDF

https://ictp.acad.ro/wp-content/uploads/2025/07/1986-Mustata-UBB-Seminar-On-the-extension-of-Lipschitz-functions.pdf

About this paper

Journal
Publisher Name
DOI
Print ISSN
Online ISSN

google scholar link

[1] Cobzas, S., Mustata, C., Norm Preserving Exctension of Convex Lipschitz Functions, J.A.T., 24 (1978), 236-244.
[2] Dunham, C.B., Chebyshev approximation with a null space, Proc. Amer. Math. Soc. 41 (1973) 557-558.
[3] Johnson, J.A., Banach Spaces of Lipschitz Functions and Vector-Valued Lipschitz Funcitons, Trans. Amer. Math. Soc. 148 (1970), 147-169.
[4] Shane, E.J., Extension of range of functions, Bull. Amer. Math. Soc. 40 (1934), 837-842.
[5] Mustata, C., Best Approximaiton and Unique Extension of Lipschitz Functions, J.A.T., 19 (1977), 222-230.
[6] Roy, A.K., Extreme Points and Linear Isometries of Banach Space of Lipschitz Functions, Canad. J. of Math. 20 (1968), 1150-1164.
[7] Singer, I., Cea mai bună aproximare în spații vectoriale normate prin elemente din subspații vectoriale, Edit. Acad. R.S. Română, București, 1967.

Paper (preprint) in HTML form

1986-Mustata-UBB-Seminar-On-the-extension-of-Lipschitz-functions

ON THE EXTENSION OF LIPSCHITZ FUNCTIONS

Costică Mustăta

  1. Introduction. Let A A AAA be a subset of the interval [ a , b ] ⊂ R [ a , b ] ⊂ R [a,b]sub R[a, b] \subset R[a,b]⊂R. A function f : A → R f : A → R f:A rarr Rf: A \rightarrow Rf:A→R is called Lipschitz if there exists L ⩾ 0 L ⩾ 0 L >= 0L \geqslant 0L⩾0 such that
(1) | f ( x ) − f ( y ) | ⩽ L ⋅ | x − y | , (1) | f ( x ) − f ( y ) | ⩽ L ⋅ | x − y | , {:(1)|f(x)-f(y)| <= L*|x-y|",":}\begin{equation*} |f(x)-f(y)| \leqslant L \cdot|x-y|, \tag{1} \end{equation*}(1)|f(x)−f(y)|⩽L⋅|x−y|,
for all x , y ∈ A x , y ∈ A x,y in Ax, y \in Ax,y∈A. The smallest number L L LLL for which the inerguality
(1) holds is called the Lipschitz norm of f f fff and is denotod by ‖ f ‖ L ‖ f ‖ L ||f||_(L)\|\mathbf{f}\|_{L}‖f‖L. The Lipschitz norm of f f fff can be calculated also by the formula
(2) ‖ f ‖ L = sup { | f ( x ) − f ( y ) | / | x − y | : x , y ∈ A , x ≠ y } ‖ f ‖ L = sup { | f ( x ) − f ( y ) | / | x − y | : x , y ∈ A , x ≠ y } ||f||_(L)=s u p{|f(x)-f(y)|//|x-y|:x,y in A,x!=y}\|f\|_{L}=\sup \{|f(x)-f(y)| /|x-y|: x, y \in A, x \neq y\}‖f‖L=sup{|f(x)−f(y)|/|x−y|:x,y∈A,x≠y}.
Denote by Lip A the set of all real valued Lipschitz func tions on A, i.e.
(3) Lip A = { P , f : A → R , f Lip A = { P , f : A → R , f quad Lip A={P,f:A rarr R,f\quad \operatorname{Lip} A=\{P, f: A \rightarrow R, fLipA={P,f:A→R,f is Lipschitz } } }\}}.
With the usual (i.e. pointwise ) operations of addition and multiplication by scalars, Lip A is a vector space.
A Lipschitz extension of f f fff to [ a , b ] [ a , b ] [a,b][a, b][a,b] is a Lipschitz function [ a , b ] → R [ a , b ] → R [a,b]rarrR[\mathrm{a}, \mathrm{b}] \rightarrow \mathrm{R}[a,b]→R such that
F | A ′ = P and ‖ F ‖ L = ‖ I ‖ L F A ′ = P  and  ‖ F ‖ L = ‖ I ‖ L F|_(A)^(')=P" and "||F||_(L)=||I||_(L)\left.F\right|_{A} ^{\prime}=P \text { and }\|F\|_{L}=\|I\|_{L}F|A′=P and ‖F‖L=‖I‖L
By a result of Mo SHANE [4] every function in Iip A A AAA has at least one Lipschitz extension in Lip [ a , b ] Lip [ a , b ] Lip[a,b]\operatorname{Lip}[\mathrm{a}, \mathrm{b}]Lip[a,b]. More exactly, the following two functions
F 1 ( x ) = sup { f ( y ) − ‖ f ‖ L ⋅ | x − y | : y ∈ A } F 1 ( x ) = sup f ( y ) − ‖ f ‖ L ⋅ | x − y | : y ∈ A F_(1)(x)=s u p{f(y)-||f||_(L)*|x-y|:y in A}F_{1}(x)=\sup \left\{f(y)-\|f\|_{L} \cdot|x-y|: y \in A\right\}F1(x)=sup{f(y)−‖f‖L⋅|x−y|:y∈A}
and.
(6) F 2 ( x ) = inf { f ( y ) + ‖ f ‖ I ′ ⋅ | x − y | : y ∈ A } (6) F 2 ( x ) = inf f ( y ) + ‖ f ‖ I ′ ⋅ | x − y | : y ∈ A {:(6)F_(2)(x)=i n f{f(y)+||f||_(I^('))*|x-y|quad:quad y inA}:}\begin{equation*} F_{2}(x)=\inf \left\{f(y)+\|f\|_{I^{\prime}} \cdot|x-y| \quad: \quad y \in \mathbb{A}\right\} \tag{6} \end{equation*}(6)F2(x)=inf{f(y)+‖f‖I′⋅|x−y|:y∈A}
are Lipschitz extensions of f f fff to [ a , b ] [ a , b ] [a,b][a, b][a,b]. Denoting by E ( f ; [ a , b ] ) E ( f ; [ a , b ] ) E(f;[a,b])E(f ;[a, b])E(f;[a,b]) the set of all Lipschitz extensions of f f fff to [ a , b ] [ a , b ] [a,b][a, b][a,b], i.e.
(7) E ( f ; [ a , b ] ) = { F ∈ Lip [ a , b ] : F | A = f E ( f ; [ a , b ] ) = F ∈ Lip [ a , b ] : F A = f E(f;[a,b])={F in Lip[a,b]:F|_(A)=f:}\mathbb{E}(f ;[a, b])=\left\{F \in \operatorname{Lip}[a, b]:\left.F\right|_{A}=f\right.E(f;[a,b])={F∈Lip[a,b]:F|A=f and ‖ F ‖ L = ‖ f ‖ L } ‖ F ‖ L = ‖ f ‖ L {:||F||_(L)=||f||_(L)}\left.\|F\|_{L}=\|f\|_{L}\right\}‖F‖L=‖f‖L}
the following assertions hold true
(a) F 1 ( x ) ≤ F ( x ) ≤ F 2 ( x ) , x ∈ [ a , b ] F 1 ( x ) ≤ F ( x ) ≤ F 2 ( x ) , x ∈ [ a , b ] quadF_(1)(x) <= F(x) <= F_(2)(x),x in[a,b]\quad F_{1}(x) \leq F(x) \leq F_{2}(x), x \in[a, b]F1(x)≤F(x)≤F2(x),x∈[a,b], for all F ∈ R ( f ; [ a , b ] ) F ∈ R ( f ; [ a , b ] ) F inR(f;[a,b])F \in \mathbb{R}(f ;[a, b])F∈R(f;[a,b]);
(b) E ( f ; [ a , b ] ) E ( f ; [ a , b ] ) quad E(f;[a,b])\quad E(f ;[a, b])E(f;[a,b]) is a convex subset of Lip [ a , b ] Lip [ a , b ] Lip[a,b]\operatorname{Lip}[a, b]Lip[a,b];
(c) The functions F 1 F 1 F_(1)\mathrm{F}_{1}F1 and F 2 F 2 F_(2)\mathrm{F}_{2}F2 are extreme points of E ( f ; [ a , b ] ) E ( f ; [ a , b ] ) E(f;[a,b])E(f ;[a, b])E(f;[a,b]).
By the definition of the Lipschitz noxm ( or by (2) ), ‖ f ‖ L = 0 ‖ f ‖ L = 0 ||f||_(L)=0\|f\|_{L}=0‖f‖L=0 if and only if f = f = f=f=f= constant and therefore ‖ ‖ ||\|‖." I I III is not actually a norm on Lip A Lip A Lip A\operatorname{Lip} ALipA but it is a norm on the space Lip 0 0 _(0){ }_{0}0 of all functions in Lip A A AAA vanishing at a fixed point x 0 ∈ A x 0 ∈ A x_(0)in Ax_{0} \in Ax0∈A. The space Lip A A A A _(A)^(A){ }_{\mathrm{A}}{ }^{\mathrm{A}}AA with the Lipschitz norm is a dual Banach space (see [3]).
2. Lipschitz extensions from finite subsets of [ a , b ] [ a , b ] [a,b][a, b][a,b].
Let C [ a , b ] C [ a , b ] C[a,b]C[a, b]C[a,b] be the space of all real valued continuous functions on [ a , b ] [ a , b ] [a,b][a, b][a,b] and let
(8) u = { x 0 , x 1 , … , x n } , a ⩽ x 0 < x 1 < … < x n ⩽ b , (8) u = x 0 , x 1 , … , x n , a ⩽ x 0 < x 1 < … < x n ⩽ b , {:(8)u={x_(0),x_(1),dots,x_(n)}quad","quad a <= x_(0) < x_(1) < dots < x_(n) <= b quad",":}\begin{equation*} u=\left\{x_{0}, x_{1}, \ldots, x_{n}\right\} \quad, \quad a \leqslant x_{0}<x_{1}<\ldots<x_{n} \leqslant b \quad, \tag{8} \end{equation*}(8)u={x0,x1,…,xn},a⩽x0<x1<…<xn⩽b,
be a linite subset of [ a , b ] [ a , b ] [a,b][\mathrm{a}, \mathrm{b}][a,b]. Then obviously, the restriction f | 11 f 11 f|_(11)\left.f\right|_{11}f|11 of a a aaa function f ∈ C [ a , b ] f ∈ C [ a , b ] f in C[a,b]f \in C[a, b]f∈C[a,b] to M M MMM is in Lip M Lip M Lip M\operatorname{Lip} MLipM and
(9) | f | M ‖ L = max { | f ( x i ) − f ( x j ) | / | x i − x j | : i , j = 0 , 1 , … n , i ≠ j } | f | M ‖ L = max f x i − f x j / x i − x j : i , j = 0 , 1 , … n , i ≠ j quad|f|_(M)||_(L)=max{|f(x_(i))-f(x_(j))|//|x_(i)-x_(j)|:i,j=0,1,dots n,i!=j}\quad|f|_{M} \|_{L}=\max \left\{\left|f\left(x_{i}\right)-f\left(x_{j}\right)\right| /\left|x_{i}-x_{j}\right|: i, j=0,1, \ldots n, i \neq j\right\}|f|M‖L=max{|f(xi)−f(xj)|/|xi−xj|:i,j=0,1,…n,i≠j} Let U : C [ a , b ] → Lip M U : C [ a , b ] → Lip M U:C[a,b]rarr Lip MU: C[a, b] \rightarrow \operatorname{Lip} MU:C[a,b]→LipM be the restriction operator, i.e.
(10)
U ( f ) = f | M , for f ∈ C [ a , b ] U ( f ) = f M ,  for  f ∈ C [ a , b ] U(f)=f|_(M)," for "f in C[a,b]U(f)=\left.f\right|_{M}, \text { for } f \in C[a, b]U(f)=f|M, for f∈C[a,b]
By the above quated result of HC SHANE, f | M f M f|_(M)\left.f\right|_{M}f|M has at least a Lipschitz extension F ∈ Lip [ a , b ] F ∈ Lip [ a , b ] F in Lip[a,b]F \in \operatorname{Lip}[a, b]F∈Lip[a,b]. Let V : Lip M → P ( Lip [ a , b ] ) V : Lip M → P ( Lip [ a , b ] ) V:Lip M rarrP(Lip[a,b])V: \operatorname{Lip} M \rightarrow \mathscr{P}(\operatorname{Lip}[a, b])V:LipM→P(Lip[a,b]) the extension operator defined by
(11) V ( g ) = E ( g ; [ a , b ] ) , g ∈ Lip M , (11) V ( g ) = E ( g ; [ a , b ] ) , g ∈ Lip M , {:(11)V(g)=E(g;[a","b])","quad g in Lip M",":}\begin{equation*} V(g)=E(g ;[a, b]), \quad g \in \operatorname{Lip} M, \tag{11} \end{equation*}(11)V(g)=E(g;[a,b]),g∈LipM,
and let W : C [ a , b ] → G ρ ( Lip [ a , b ] ) W : C [ a , b ] → G ρ ( Lip [ a , b ] ) W:C[a,b]rarrG^(rho)(Lip[a,b])W: C[a, b] \rightarrow \mathcal{G}^{\rho}(\operatorname{Lip}[a, b])W:C[a,b]→Gρ(Lip[a,b]) be the composition of U U UUU and V V VVV
(12) W = V 0 U (12) W = V 0 U {:(12)W=V_(0)U:}\begin{equation*} W=V_{0} U \tag{12} \end{equation*}(12)W=V0U
In general W W WWW is a multivalued operator (point to set) . A function f ∈ C [ a , b ] f ∈ C [ a , b ] f in C[a,b]f \in C[a, b]f∈C[a,b] such that f ∈ W ( f ) f ∈ W ( f ) f in W(f)f \in W(f)f∈W(f) is called a fix point of W W WWW.
Obviously, the set of fix points of the operator W W WWW is non-void.
Indeed, if g ∈ Lip M g ∈ Lip M g in Lip Mg \in \operatorname{Lip} Mg∈LipM and f f fff is a Lipschitz extension of G G GGG to [ a , b ] [ a , b ] [a,b][a, b][a,b], then f ∈ W ( f ) f ∈ W ( f ) f in W(f)f \in W(f)f∈W(f). The fix points of the operator W W WWW are characterized in Theorem 2.1 below.
For x , y ∈ [ a , b ] , x ≠ y and f ∈ C [ a , b ] , put  For  x , y ∈ [ a , b ] , x ≠ y  and  f ∈ C [ a , b ] , put  " For "x,y in[a,b],x!=y" and "f in C[a,b]", put "\text { For } x, y \in[a, b], x \neq y \text { and } f \in C[a, b] \text {, put } For x,y∈[a,b],x≠y and f∈C[a,b], put 
(13) [ x , y ; f ] = ( f ( x ) − f ( y ) ) / ( x − y ) (13) [ x , y ; f ] = ( f ( x ) − f ( y ) ) / ( x − y ) {:(13)[x","y;f]=(f(x)-f(y))//(x-y):}\begin{equation*} [x, y ; f]=(f(x)-f(y)) /(x-y) \tag{13} \end{equation*}(13)[x,y;f]=(f(x)−f(y))/(x−y)
and
(14) f ( x , y ; f ) ( t ) = [ x , y ; f ] ⋅ ( t − x ) + f ( x ) , t ∈ [ a , b ] (14) f ( x , y ; f ) ( t ) = [ x , y ; f ] ⋅ ( t − x ) + f ( x ) , t ∈ [ a , b ] {:(14)f(x","y;f)(t)=[x","y;f]*(t-x)+f(x)","quad t in[a","b]:}\begin{equation*} f(x, y ; f)(t)=[x, y ; f] \cdot(t-x)+f(x), \quad t \in[a, b] \tag{14} \end{equation*}(14)f(x,y;f)(t)=[x,y;f]⋅(t−x)+f(x),t∈[a,b]
2.1 THEORTM, Let f ∈ C [ a , b ] f ∈ C [ a , b ] finC[a,b]\mathrm{f} \in \mathrm{C}[\mathrm{a}, \mathrm{b}]f∈C[a,b] and let M be the set ( e ).
Then f ∈ W ( f ) f ∈ W ( f ) f in W(f)f \in W(f)f∈W(f) if and only if there exists an index k ∈ { 0 , 1 , … , n − 1 } k ∈ { 0 , 1 , … , n − 1 } k in{0,1,dots,n-1}k \in\{0,1, \ldots, n-1\}k∈{0,1,…,n−1} such that
(15) sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ∉ y } = | [ x k , x k + 1 ; f ] | sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ∉ y } = x k , x k + 1 ; f s u p{|[x,y;f]|:quad x,y in[a,b],quad x!in y}=|[x_(k),x_(k+1);f]|\sup \{|[x, y ; f]|: \quad x, y \in[a, b], \quad x \notin y\}=\left|\left[x_{k}, x_{k+1} ; f\right]\right|sup{|[x,y;f]|:x,y∈[a,b],x∉y}=|[xk,xk+1;f]|.
Proof. If f ∈ W ( f ) f ∈ W ( f ) f in W(f)f \in W(f)f∈W(f) then f ∈ E ( f | M ; [ a , b ] ) f ∈ E f M ; [ a , b ] f in E(f|_(M);[a,b])f \in E\left(\left.f\right|_{M} ;[a, b]\right)f∈E(f|M;[a,b]) and
‖ I ‖ L = ‖ f ‖ M ‖ L = max { | [ x j , x j + 1 ; f ] | : j = 0 , 1 , … , n − 1 } ‖ I ‖ L = ‖ f ‖ M ‖ L = max x j , x j + 1 ; f : j = 0 , 1 , … , n − 1 ||I||_(L)=||f||_(M)||_(L)=max{|[x_(j),x_(j+1);f]|:j=0,1,dots,n-1}\|I\|_{L}=\|f\|_{M} \|_{L}=\max \left\{\left|\left[x_{j}, x_{j+1} ; f\right]\right|: j=0,1, \ldots, n-1\right\}‖I‖L=‖f‖M‖L=max{|[xj,xj+1;f]|:j=0,1,…,n−1}
so that, there exists k ∈ { 0 , 1 , … , n − 1 } k ∈ { 0 , 1 , … , n − 1 } k in{0,1,dots,n-1}k \in\{0,1, \ldots, n-1\}k∈{0,1,…,n−1} such that ‖ f ‖ M ‖ L = ‖ f ‖ M ‖ L = ||f||_(M)||_(L)=\|f\|_{M} \|_{L}=‖f‖M‖L=
| [ x k , x k + 1 ; f ] | x k , x k + 1 ; f |[x_(k),x_(k+1);f]|\left|\left[x_{k}, x_{k+1} ; f\right]\right||[xk,xk+1;f]|, and the relation (15) holds.
Conversely, if the relation (15) holds for a index k ∈ { 0 , 1 , … k ∈ { 0 , 1 , … k in{0,1,dotsk \in\{0,1, \ldotsk∈{0,1,…, n − 1 } n − 1 } n-1}n-1\}n−1}, then
| [ x k , x k + 1 ; f ] | = sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ≠ y } ⩾ x k , x k + 1 ; f = sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ≠ y } ⩾ |[x_(k),x_(k+1);f]|=s u p{|[x,y;f]|:x,y in[a,b],x!=y} >=\left|\left[x_{k}, x_{k+1} ; f\right]\right|=\sup \{|[x, y ; f]|: x, y \in[a, b], x \neq y\} \geqslant|[xk,xk+1;f]|=sup{|[x,y;f]|:x,y∈[a,b],x≠y}⩾
⩾ max { | [ x j , x j + 1 ; f ] | : j = 0 , 1 , … , n − 1 } = ‖ f | M ‖ L ⩾ max x j , x j + 1 ; f : j = 0 , 1 , … , n − 1 = f M L >= max{|[x_(j),x_(j+1);f]|:j=0,1,dots,n-1}=||f|_(M)||_(L)\geqslant \max \left\{\left|\left[x_{j}, x_{j+1} ; f\right]\right|: j=0,1, \ldots, n-1\right\}=\left\|\left.f\right|_{M}\right\|_{L}⩾max{|[xj,xj+1;f]|:j=0,1,…,n−1}=‖f|M‖L.
Therefore, | f | ‖ I = | [ x k , x k + 1 ; f ] | = ‖ f ‖ L | f | I = x k , x k + 1 ; f = f ‖ L |f|||_(I)=|[x_(k),x_(k+1);f]|=||f||_(L)|f|\left\|_{I}=\left|\left[x_{k}, x_{k+1} ; f\right]\right|=\right\| f \|_{L}|f|‖I=|[xk,xk+1;f]|=‖f‖L, which shows that f ∈ W ( f ) f ∈ W ( f ) f in W(f)f \in W(f)f∈W(f).
Theoret 2.1 has some corollaries.
2.2 COROITARY . If the relation (15) holds for an index k ∈ { 0 , 1 , … , n − 1 } k ∈ { 0 , 1 , … , n − 1 } k in{0,1,dots,n-1}k \in\{0,1, \ldots, n-1\}k∈{0,1,…,n−1} then
f ( x ) = L ( x k , x k + 1 ; f ) ( x ) for all x ∈ [ x k , x k + 1 ] f ( x ) = L x k , x k + 1 ; f ( x )  for all  x ∈ x k , x k + 1 f(x)=L(x_(k),x_(k+1);f)(x)" for all "x in[x_(k),x_(k+1)]f(x)=\mathcal{L}\left(x_{k}, x_{k+1} ; f\right)(x) \text { for all } x \in\left[x_{k}, x_{k+1}\right]f(x)=L(xk,xk+1;f)(x) for all x∈[xk,xk+1]
Proof. If f ( x ′ ) ≠ ℓ ( x k , x k + 1 ; f ) ( x ′ ) f x ′ ≠ ℓ x k , x k + 1 ; f x ′ f(x^('))!=ℓ(x_(k),x_(k+1);f)(x^('))f\left(x^{\prime}\right) \neq \ell\left(x_{k}, x_{k+1} ; f\right)\left(x^{\prime}\right)f(x′)≠ℓ(xk,xk+1;f)(x′), for an x ′ ∈ ( x k , x k + 1 ) x ′ ∈ x k , x k + 1 x^(')in(x_(k),x_(k+1))x^{\prime} \in\left(x_{k}, x_{k+1}\right)x′∈(xk,xk+1)
then
max { | [ x k , x ′ ; f ] | , | [ x ′ , x k + 1 ; f ] | } > | [ x k , x k + 1 ; f ] | = ⋅ = sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ≠ y } max x k , x ′ ; f , x ′ , x k + 1 ; f > x k , x k + 1 ; f = ⋅ = sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ≠ y } {:[max{|[x_(k),x^(');f]|,:}{:|[x^('),x_(k+1);f]|} > |[x_(k),x_(k+1);f]|=],[*=s u p{|[x","y;f]|:x","y in[a","b]","x!=y}]:}\begin{aligned} \max \left\{\left|\left[x_{k}, x^{\prime} ; f\right]\right|,\right. & \left.\left|\left[x^{\prime}, x_{k+1} ; f\right]\right|\right\}>\left|\left[x_{k}, x_{k+1} ; f\right]\right|= \\ \cdot & =\sup \{|[x, y ; f]|: x, y \in[a, b], x \neq y\} \end{aligned}max{|[xk,x′;f]|,|[x′,xk+1;f]|}>|[xk,xk+1;f]|=⋅=sup{|[x,y;f]|:x,y∈[a,b],x≠y}
which is a contradiction.
2.3 COROILARY. If f ∈ W ( f ) f ∈ W ( f ) f in W(f)f \in W(f)f∈W(f) then f ( x ) = F ( x ) , x ∈ [ x k , x k + 1 ] f ( x ) = F ( x ) , x ∈ x k , x k + 1 f(x)=F(x),x in[x_(k),x_(k+1)]f(x)=F(x), x \in\left[x_{k}, x_{k+1}\right]f(x)=F(x),x∈[xk,xk+1], for 311. F ∈ W ( f ) F ∈ W ( f ) F in W(f)F \in W(f)F∈W(f), where k ∈ { 0 , 1 , … , n − 1 } k ∈ { 0 , 1 , … , n − 1 } k in{0,1,dots,n-1}k \in\{0,1, \ldots, n-1\}k∈{0,1,…,n−1} is the index for which relation (25) is true.
Proof. If F ∈ W ( ± ) F ∈ W ( ± ) F in W(+-)F \in W( \pm)F∈W(±) then
sup { | [ x , y ; F ] | : x , y ∈ [ a , b ] , x ≠ y } = ‖ F ‖ I = ‖ f ‖ M ‖ L =∣ [ x k , x k + 1 ; F ‖ F ∣ sup { | [ x , y ; F ] | : x , y ∈ [ a , b ] , x ≠ y } = ‖ F ‖ I = ‖ f ‖ M ‖ L =∣ x k , x k + 1 ; F ‖ F ∣ s u p{|[x,y;F]|:quadx,yin[a,b],x!=y}=||F||_(I)=||f||_(M)||_(L)=∣[x_(k),x_(k+1);F||^(F)∣:}\sup \{|[\mathrm{x}, \mathrm{y} ; \mathrm{F}]|: \quad \mathrm{x}, \mathrm{y} \in[\mathrm{a}, \mathrm{b}], \mathrm{x} \neq \mathrm{y}\}=\|\mathrm{F}\|_{\mathrm{I}}=\|\mathrm{f}\|_{\mathrm{M}} \|_{\mathrm{L}}=\mid\left[\mathrm{x}_{\mathrm{k}}, \mathrm{x}_{\mathrm{k}+1} ; \mathrm{F} \|^{\mathrm{F}} \mid\right.sup{|[x,y;F]|:x,y∈[a,b],x≠y}=‖F‖I=‖f‖M‖L=∣[xk,xk+1;F‖F∣
and Corollary - 2.3 follows from Corollary 2.2 .
2.4 COROTLARY. If a = x 0 a = x 0 a=x_(0)a=x_{0}a=x0 and x n = b x n = b x_(n)=bx_{n}=bxn=b, then
(a) sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ≠ y } = | [ x 0 , x n ; f ] | sup { | [ x , y ; f ] | : x , y ∈ [ a , b ] , x ≠ y } = x 0 , x n ; f s u p{|[x,y;f]|:x,y in[a,b],x!=y}=|[x_(0),x_(n);f]|\sup \{|[x, y ; f]|: x, y \in[a, b], x \neq y\}=\left|\left[x_{0}, x_{n} ; f\right]\right|sup{|[x,y;f]|:x,y∈[a,b],x≠y}=|[x0,xn;f]|, implies f ( x ) = f ( x 0 , x n ; f ) ( x ) f ( x ) = f x 0 , x n ; f ( x ) f(x)=f(x_(0),x_(n);f)(x)f(x)=f\left(x_{0}, x_{n} ; f\right)(x)f(x)=f(x0,xn;f)(x), for all x ∈ [ a , b ] x ∈ [ a , b ] x in[a,b]x \in[a, b]x∈[a,b], and
(b) sup { | [ x , y ; e ] | : x , y ∈ { a , b ] , x ≠ y } = | [ x k , x k + 1 ; f ] | sup { | [ x , y ; e ] | : x , y ∈ { a , b ] , x ≠ y } = x k , x k + 1 ; f s u p{|[x,y;e]|:x,y in{a,b],x!=y}=|[x_(k),x_(k+1);f]|\sup \{|[x, y ; e]|: x, y \in\{a, b], x \neq y\}=\left|\left[x_{k}, x_{k+1} ; f\right]\right|sup{|[x,y;e]|:x,y∈{a,b],x≠y}=|[xk,xk+1;f]|, k = 0 , 1 , … , n − 1 k = 0 , 1 , … , n − 1 k=0,1,dots,n-1k=0,1, \ldots, n-1k=0,1,…,n−1, implies f ( x ) = f ( x k , x k + 1 ; f ) ( x ) f ( x ) = f x k , x k + 1 ; f ( x ) f(x)=f(x_(k),x_(k+1);f)(x)f(x)=f\left(x_{k}, x_{k+1} ; f\right)(x)f(x)=f(xk,xk+1;f)(x), x ∈ [ x k , x k + 1 ] , k = 0 , 1 , … , n − 1 x ∈ x k , x k + 1 , k = 0 , 1 , … , n − 1 x in[x_(k),x_(k+1)]quad,k=0,1,dots,n-1x \in\left[x_{k}, x_{k+1}\right] \quad, k=0,1, \ldots, n-1x∈[xk,xk+1],k=0,1,…,n−1.
Combining Corollaries 2.3 and 2.4 it follows
2.5 COROLLARY. If f ∈ Iip [ a , b ] , x o = a , x n = b f ∈ Iip [ a , b ] , x o = a , x n = b f in Iip[a,b],x_(o)=a,x_(n)=bf \in \operatorname{Iip}[a, b], x_{o}=a, x_{n}=bf∈Iip[a,b],xo=a,xn=b and ‖ f ‖ L ==∣ [ x k , x k + 1 ; f ] , k = 0 , 1 , … , n − 1 ‖ f ‖ L ==∣ x k , x k + 1 ; f , k = 0 , 1 , … , n − 1 ||f||_(L)==∣[x_(k),x_(k+1);f],k=0,1,dots,n-1\|f\|_{L}= =\mid\left[x_{k}, x_{k+1} ; f\right], k=0,1, \ldots, n-1‖f‖L==∣[xk,xk+1;f],k=0,1,…,n−1, then W ( f ) = { f } W ( f ) = { f } W(f)={f}W(f)=\{f\}W(f)={f}.
3. Faces and extreme points.
Let X X XXX be a normed space and B ( X ) = { X G X : ‖ x ‖ ≤ 1 } B ( X ) = { X G X : ‖ x ‖ ≤ 1 } B(X)={XGX:||x|| <= 1}B(X)=\{X G X:\|x\| \leq 1\}B(X)={XGX:‖x‖≤1} its closed unit ball. A subset A ⊆ B ( X ) A ⊆ B ( X ) A sube B(X)A \subseteq B(X)A⊆B(X) is called an extremal subset (a face) of B ( X ) B ( X ) B(X)B(X)B(X) if α f 1 + ( 1 − α ) f 2 ∈ A α f 1 + ( 1 − α ) f 2 ∈ A alphaf_(1)+(1-alpha)f_(2)in A\alpha f_{1}+(1-\alpha) f_{2} \in Aαf1+(1−α)f2∈A for f 1 , f 2 ∈ B ( X ) f 1 , f 2 ∈ B ( X ) f_(1),f_(2)in B(X)f_{1}, f_{2} \in B(X)f1,f2∈B(X) and a number α , 0 < α < 1 α , 0 < α < 1 alpha,0 < alpha < 1\alpha, 0<\alpha<1α,0<α<1, implies f 1 , f 2 ∈ A f 1 , f 2 ∈ A f_(1),f_(2)in Af_{1}, f_{2} \in Af1,f2∈A. If A A AAA contains exactly one point f f fff, then f f fff is called an extreme point of B ( X ) B ( X ) B(X)B(X)B(X).
Let M M MMM be a finite set of real numbers
and
M = { x 0 , x 1 , … , x n } , x 0 < x 1 < … < x n M = x 0 , x 1 , … , x n , x 0 < x 1 < … < x n M={x_(0),x_(1),dots,x_(n)}quad,quadx_(0) < x_(1) < dots < x_(n)M=\left\{x_{0}, x_{1}, \ldots, x_{n}\right\} \quad, \quad x_{0}<x_{1}<\ldots<x_{n}M={x0,x1,…,xn},x0<x1<…<xn
Lip 0 M ′ = { f , Lip 0 M ′ = f , Lip_(0)M^(')={f,quad:}\operatorname{Lip}_{0} M^{\prime}=\left\{f, \quad\right.Lip0M′={f, f : M → R , f : M → R , f :M rarr R,quad f: M \rightarrow R, \quad f:M→R,f is Lipschitz and f ( x 0 ) = 0 } f x 0 = 0 {:f(x_(0))=0}\left.f\left(x_{0}\right)=0\right\}f(x0)=0}.
3.1 THEOREM. The function f ∈ B ( f ∈ B f in B(:}f \in B\left(\right.f∈B( Lip 0 M ) 0 M {:_(0)M)\left._{0} M\right)0M) is an extreme point of B ( Lip 0 M ) B Lip 0 M B(Lip_(0)M)B\left(\operatorname{Lip}_{0} M\right)B(Lip0M) if and only if
(16) ‖ f ‖ L = 1 =∣ [ x k , x k + 1 ; f ] ] (16) ‖ f ‖ L = 1 =∣ x k , x k + 1 ; f {:(16){:||f||_(L)=1=∣[x_(k),x_(k+1);f]]:}\begin{equation*} \left.\|f\|_{L}=1=\mid\left[x_{k}, x_{k+1} ; f\right]\right] \tag{16} \end{equation*}(16)‖f‖L=1=∣[xk,xk+1;f]]
for k = 0 , 1 , … , n − 1 k = 0 , 1 , … , n − 1 k=0,1,dots,n-1k=0,1, \ldots, n-1k=0,1,…,n−1.
Proof. Suppose that relation (16) do not hold and let k ∈ { 0 , 1 , … , n − 1 } k ∈ { 0 , 1 , … , n − 1 } kin{0,1,dots,n-1}\mathbb{k} \in\{0,1, \ldots, n-1\}k∈{0,1,…,n−1} and ε > 0 ε > 0 epsi > 0\varepsilon>0ε>0 be such that
| [ x k , x k + 1 ; f ] | = 1 − ε . x k , x k + 1 ; f = 1 − ε . |[x_(k),x_(k+1);f]|=1-epsi.\left|\left[x_{k}, x_{k+1} ; f\right]\right|=1-\varepsilon .|[xk,xk+1;f]|=1−ε.
Let
f 1 ( x ) = { f ( x ) , x ∈ { x 0 , x 1 , … , x k } f ( x ) + δ , x ∈ { x k + 1 , … , x n } f 1 ( x ) = f ( x ) ,      x ∈ x 0 , x 1 , … , x k f ( x ) + δ ,      x ∈ x k + 1 , … , x n f_(1)(x)={[f(x)",",x in{x_(0),x_(1),dots,x_(k)}],[f(x)+delta",",x in{x_(k+1),dots,x_(n)}]:}f_{1}(x)= \begin{cases}f(x), & x \in\left\{x_{0}, x_{1}, \ldots, x_{k}\right\} \\ f(x)+\delta, & x \in\left\{x_{k+1}, \ldots, x_{n}\right\}\end{cases}f1(x)={f(x),x∈{x0,x1,…,xk}f(x)+δ,x∈{xk+1,…,xn}
and
f 2 ( x ) = { f ( x ) , x ∈ { x 0 , x 1 , … , x k } f ( x ) − δ , x ∈ { x k + 1 , … , x n } f 2 ( x ) = f ( x ) ,      x ∈ x 0 , x 1 , … , x k f ( x ) − δ ,      x ∈ x k + 1 , … , x n f_(2)(x)={[f(x)",",x in{x_(0),x_(1),dots,x_(k)}],[f(x)-delta",",x in{x_(k+1),dots,x_(n)}]:}f_{2}(x)= \begin{cases}f(x), & x \in\left\{x_{0}, x_{1}, \ldots, x_{k}\right\} \\ f(x)-\delta, & x \in\left\{x_{k+1}, \ldots, x_{n}\right\}\end{cases}f2(x)={f(x),x∈{x0,x1,…,xk}f(x)−δ,x∈{xk+1,…,xn}
where δ = ( ε / 2 ) ( x k + 1 − x k ) δ = ( ε / 2 ) x k + 1 − x k delta=(epsi//2)(x_(k+1)-x_(k))\delta=(\varepsilon / 2)\left(x_{k+1}-x_{k}\right)δ=(ε/2)(xk+1−xk).
Because
| f ( x k + 1 ) + δ − f ( x k ) | x k + 1 − x k ⩽ 1 − ε / 2 < 1 and | f ( x k + 1 ) − δ − f ( x k ) | x k + 1 − x k ⩽ 1 − ε / 2 < 1 , it follows that f x k + 1 + δ − f x k x k + 1 − x k ⩽ 1 − ε / 2 < 1  and  f x k + 1 − δ − f x k x k + 1 − x k ⩽ 1 − ε / 2 < 1 , it follows that  {:[(|f(x_(k+1))+delta-f(x_(k))|)/(x_(k+1)-x_(k)) <= 1-epsi//2 < 1quad" and "],[(|f(x_(k+1))-delta-f(x_(k))|)/(x_(k+1)-x_(k)) <= 1-epsi//2 < 1", it follows that "]:}\begin{aligned} & \frac{\left|f\left(x_{k+1}\right)+\delta-f\left(x_{k}\right)\right|}{x_{k+1}-x_{k}} \leqslant 1-\varepsilon / 2<1 \quad \text { and } \\ & \frac{\left|f\left(x_{k+1}\right)-\delta-f\left(x_{k}\right)\right|}{x_{k+1}-x_{k}} \leqslant 1-\varepsilon / 2<1 \text {, it follows that } \end{aligned}|f(xk+1)+δ−f(xk)|xk+1−xk⩽1−ε/2<1 and |f(xk+1)−δ−f(xk)|xk+1−xk⩽1−ε/2<1, it follows that 
‖ f 1 ‖ L = 1 = ‖ f 2 ‖ J f 1 L = 1 = f 2 J ||f_(1)||_(L)=1=||f_(2)||_(J)\left\|f_{1}\right\|_{L}=1=\left\|f_{2}\right\|_{J}‖f1‖L=1=‖f2‖J. But f = ( 1 / 2 ) ( f 1 + f 2 ) f = ( 1 / 2 ) f 1 + f 2 f=(1//2)(f_(1)+f_(2))f=(1 / 2)\left(f_{1}+f_{2}\right)f=(1/2)(f1+f2), so that f f fff is not an extreme point of B ( Lip 0 M ) B Lip 0 M B(Lip_(0)M)B\left(\operatorname{Lip}_{0} M\right)B(Lip0M).
Suppose now that condition (16) is fulfilled and there exist two functions g 1 , g 2 ∈ B ( Lip 0 M ) g 1 , g 2 ∈ B Lip 0 M g_(1),g_(2)in B(Lip_(0)M)g_{1}, g_{2} \in B\left(\operatorname{Lip}_{0} M\right)g1,g2∈B(Lip0M) such that g 1 ≠ f ≠ g 2 g 1 ≠ f ≠ g 2 g_(1)!=f!=g_(2)g_{1} \neq f \neq g_{2}g1≠f≠g2 and f = ( 1 / 2 ) ( g 1 + g 2 ) f = ( 1 / 2 ) g 1 + g 2 f=(1//2)(g_(1)+g_(2))f=(1 / 2)\left(g_{1}+g_{2}\right)f=(1/2)(g1+g2). Let x i x i x_(i)x_{i}xi be the smallest element of M M MMM for which g 1 ( x i ) ≠ f ( x i ) g 1 x i ≠ f x i g_(1)(x_(i))!=f(x_(i))g_{1}\left(x_{i}\right) \neq f\left(x_{i}\right)g1(xi)≠f(xi). As g 1 ( x 0 ) = f ( x 0 ) = 0 g 1 x 0 = f x 0 = 0 g_(1)(x_(0))=f(x_(0))=0g_{1}\left(x_{0}\right)=f\left(x_{0}\right)=0g1(x0)=f(x0)=0 and g 2 ( x 0 ) = 0 == f ( x 0 ) g 2 x 0 = 0 == f x 0 g_(2)(x_(0))=0==f(x_(0))g_{2}\left(x_{0}\right)=0= =f\left(x_{0}\right)g2(x0)=0==f(x0), it follows i ⩾ 1 i ⩾ 1 i >= 1i \geqslant 1i⩾1.
Case I. f ( x i ) > f ( x i − 1 ) f x i > f x i − 1 f(x_(i)) > f(x_(i-1))f\left(x_{i}\right)>f\left(x_{i-1}\right)f(xi)>f(xi−1) and g 1 ( x i ) > f ( x i ) g 1 x i > f x i g_(1)(x_(i)) > f(x_(i))g_{1}\left(x_{i}\right)>f\left(x_{i}\right)g1(xi)>f(xi). In this case
g ( x i ) − g 1 ( x i − 1 ) x i − x i − 1 > f ( x i ) − f ( x i − 1 ) x i − x i − 1 = 1 g x i − g 1 x i − 1 x i − x i − 1 > f x i − f x i − 1 x i − x i − 1 = 1 (g(x_(i))-g_(1)(x_(i-1)))/(x_(i)-x_(i-1)) > (f(x_(i))-f(x_(i-1)))/(x_(i)-x_(i-1))=1\frac{g\left(x_{i}\right)-g_{1}\left(x_{i-1}\right)}{x_{i}-x_{i-1}}>\frac{f\left(x_{i}\right)-f\left(x_{i-1}\right)}{x_{i}-x_{i-1}}=1g(xi)−g1(xi−1)xi−xi−1>f(xi)−f(xi−1)xi−xi−1=1
which implies ‖ g 1 ‖ L > 1 g 1 L > 1 ||g_(1)||_(L) > 1\left\|g_{1}\right\|_{L}>1‖g1‖L>1, i.e. g 1 ∉ B ( Lip 0 M ) g 1 ∉ B Lip 0 M quadg_(1)!in B(Lip_(0)M)\quad g_{1} \notin B\left(\operatorname{Lip}_{0} M\right)g1∉B(Lip0M).
‖ E 1 ‖ L > g 1 ∉ B ( Lip M ) E 1 L > g 1 ∉ B ( Lip M ) ||E_(1)||_(L) > g_(1)!in B(Lip M)\left\|\mathrm{E}_{1}\right\|_{L}>\mathrm{g}_{1} \notin B(\operatorname{Lip} M)‖E1‖L>g1∉B(LipM). . . . . . .
Case II. f ( x i ) > f ( x i − 1 ) f x i > f x i − 1 f(x_(i)) > f(x_(i-1))f\left(x_{i}\right)>f\left(x_{i-1}\right)f(xi)>f(xi−1) and g 1 ( x i ) < f ( x i ) g 1 x i < f x i g_(1)(x_(i)) < f(x_(i))g_{1}\left(x_{i}\right)<f\left(x_{i}\right)g1(xi)<f(xi). In this case g 2 ( x 1 ) = 2 f ( x 1 ) − g 1 ( x i ) > f ( x i ) g 2 x 1 = 2 f x 1 − g 1 x i > f x i g_(2)(x_(1))=2f(x_(1))-g_(1)(x_(i)) > f(x_(i))g_{2}\left(x_{1}\right)=2 f\left(x_{1}\right)-g_{1}\left(x_{i}\right)>f\left(x_{i}\right)g2(x1)=2f(x1)−g1(xi)>f(xi) and
g 2 ( x i ) − g 2 ( x i − 1 ) x i − x i − 1 > f ( x i ) − [ 2 f ( x i − 1 ) − g 1 ( x i − 1 ) ] x i − x i − 1 = = f ( x i ) − f ( x i − 1 ) x i − x i − 1 = 1 g 2 x i − g 2 x i − 1 x i − x i − 1 > f x i − 2 f x i − 1 − g 1 x i − 1 x i − x i − 1 = = f x i − f x i − 1 x i − x i − 1 = 1 {:[(g_(2)(x_(i))-g_(2)(x_(i-1)))/(x_(i)-x_(i-1)) > (f(x_(i))-[2f(x_(i-1))-g_(1)(x_(i-1))])/(x_(i)-x_(i-1))=],[=(f(x_(i))-f(x_(i-1)))/(x_(i)-x_(i-1))=1]:}\begin{aligned} \frac{g_{2}\left(x_{i}\right)-g_{2}\left(x_{i-1}\right)}{x_{i}-x_{i-1}}> & \frac{f\left(x_{i}\right)-\left[2 f\left(x_{i-1}\right)-g_{1}\left(x_{i-1}\right)\right]}{x_{i}-x_{i-1}}= \\ & =\frac{f\left(x_{i}\right)-f\left(x_{i-1}\right)}{x_{i}-x_{i-1}}=1 \end{aligned}g2(xi)−g2(xi−1)xi−xi−1>f(xi)−[2f(xi−1)−g1(xi−1)]xi−xi−1==f(xi)−f(xi−1)xi−xi−1=1
so that ‖ g 2 ‖ L > 1 g 2 L > 1 ||g_(2)||_(L) > 1\left\|g_{2}\right\|_{L}>1‖g2‖L>1, i.e. g 2 ∉ B ( Iip p 0 M ) g 2 ∉ B Iip p 0 M g_(2)!in B(Iipp_(0)M)g_{2} \notin B\left(\operatorname{Iip} p_{0} M\right)g2∉B(Iipp0M).
In the remaining cases, i.e. f ( x i ) < f ( x i − 1 ) f x i < f x i − 1 f(x_(i)) < f(x_(i-1))f\left(x_{i}\right)<f\left(x_{i-1}\right)f(xi)<f(xi−1) and g 1 ( x i ) > f ( x i ) g 1 x i > f x i g_(1)(x_(i)) > f(x_(i))g_{1}\left(x_{i}\right)>f\left(x_{i}\right)g1(xi)>f(xi) or f ( x i ) < f ( x i − 1 ) f x i < f x i − 1 f(x_(i)) < f(x_(i-1))f\left(x_{i}\right)<f\left(x_{i-1}\right)f(xi)<f(xi−1) and g 1 ( x i ) < f ( x i ) g 1 x i < f x i g_(1)(x_(i)) < f(x_(i))g_{1}\left(x_{i}\right)<f\left(x_{i}\right)g1(xi)<f(xi), we have similarly ‖ g 2 ‖ L > 1 g 2 L > 1 ||g_(2)||_(L) > 1\left\|g_{2}\right\|_{L}>1‖g2‖L>1, respectively ‖ g 1 ‖ I > 1 g 1 I > 1 ||g_(1)||_(I) > 1\left\|g_{1}\right\|_{I}>1‖g1‖I>1. The obtained contradictions show that f f fff must be an extreme point of B ( Lip 0 M ) B Lip 0 M B(Lip_(0)M)B\left(\operatorname{Lip}_{0} M\right)B(Lip0M).
3.2 Remark. Taking in (16) all possible signs, it follows that the unit ball of the space Lip 0 M 0 M _(0)M{ }_{0} M0M has exactly 2 n + 1 2 n + 1 2^(n+1)2^{n+1}2n+1 extreme points.
3.3 COROLLARY. Let f ∈ Lip [ a , b ] , M = { x 0 , x 1 , … , x n } f ∈ Lip [ a , b ] , M = x 0 , x 1 , … , x n fin Lip[a,b],M={x_(0),x_(1),dots,x_(n)}\mathrm{f} \in \operatorname{\operatorname {Lip}}[\mathrm{a}, \mathrm{b}], \mathrm{M}=\left\{\mathrm{x}_{0}, \mathrm{x}_{1}, \ldots, \mathrm{x}_{\mathrm{n}}\right\}f∈Lip[a,b],M={x0,x1,…,xn}, a = x 0 < x 1 < … < x n = b a = x 0 < x 1 < … < x n = b a=x_(0) < x_(1) < dots < x_(n)=ba=x_{0}<x_{1}<\ldots<x_{n}=ba=x0<x1<…<xn=b, and f ( x 0 ) = 0 f x 0 = 0 f(x_(0))=0f\left(x_{0}\right)=0f(x0)=0. If
‖ f ‖ L = I = | [ x k , x k + 1 ; f ] | ‖ f ‖ L = I = x k , x k + 1 ; f ||f||_(L)=I=|[x_(k),x_(k+1);f]|\|f\|_{L}=I=\left|\left[x_{k}, x_{k+1} ; f\right]\right|‖f‖L=I=|[xk,xk+1;f]|
for all k = 0 , 1 , … , n − 1 k = 0 , 1 , … , n − 1 k=0,1,dots,n-1k=0,1, \ldots, n-1k=0,1,…,n−1, then f f fff is an extreme point of the unit ball of Lip o [ a , b ] Lip o [ a , b ] Lip_(o)[a,b]\operatorname{Lip}_{\mathrm{o}}[\mathrm{a}, \mathrm{b}]Lipo[a,b].
3.4 Remark. If f f fff is an extreme point of B ( Lip 0 M ) B Lip 0 M B(Lip_(0)M)B\left(\operatorname{Lip}_{0} M\right)B(Lip0M) and the set M M MMM is as in Corollary 3.3 , then the unique Lipschitz extension F F FFF of f , F ( x ) = sup { f ( x i ) − | x − x i | : i = 0 , 1 , … , n } = inf { f ′ ( x i ) + | x − x i | : i = 0 , 1 , … , n } , x ∈ [ a , b ] f , F ( x ) = sup f x i − x − x i : i = 0 , 1 , … , n = inf f ′ x i + x − x i : i = 0 , 1 , … , n , x ∈ [ a , b ] f,F(x)=s u p{f(x_(i))-|x-x_(i)|:i=0,1,dots,n}=i n f{f^(')(x_(i))+|x-x_(i)|:i=0,1,dots,n},x in[a,b]f, F(x)=\sup \left\{f\left(x_{i}\right)-\left|x-x_{i}\right|: i=0,1, \ldots, n\right\} =\inf \left\{f^{\prime}\left(x_{i}\right)+\left|x-x_{i}\right|: i=0,1, \ldots, n\right\}, x \in[a, b]f,F(x)=sup{f(xi)−|x−xi|:i=0,1,…,n}=inf{f′(xi)+|x−xi|:i=0,1,…,n},x∈[a,b], is an extreme point of B ( Lip 0 [ a , b ] ) B Lip 0 [ a , b ] B(Lip_(0)[a,b])B\left(\operatorname{Lip}_{0}[a, b]\right)B(Lip0[a,b]).

4. Best approximation of Lipschitz functions.

Let
M = { x 0 , x 1 , … , x n } , a ≤ x 0 < x 1 < … < x n ≤ b M = x 0 , x 1 , … , x n , a ≤ x 0 < x 1 < … < x n ≤ b M={x_(0),x_(1),dots,x_(n)}quad,quad a <= x_(0) < x_(1) < dots < x_(n) <= bM=\left\{x_{0}, x_{1}, \ldots, x_{n}\right\} \quad, \quad a \leq x_{0}<x_{1}<\ldots<x_{n} \leq bM={x0,x1,…,xn},a≤x0<x1<…<xn≤b
and let
(17) Lip 0 [ a , b ] = { f , f : [ a , b ] → R , f Lip 0 [ a , b ] = f , f : [ a , b ] → R , f quadLip_(0)[a,b]={f,f:[a,b]rarr R,f:}\quad \operatorname{Lip}_{0}[a, b]=\left\{f, f:[a, b] \rightarrow R, f\right.Lip0[a,b]={f,f:[a,b]→R,f is Lipschitz and f ( x 0 ) = 0 } f x 0 = 0 {:f(x_(0))=0}\left.f\left(x_{0}\right)=0\right\}f(x0)=0}. Let also
M ⊥ = { f ∈ Li p o [ a , b ] ; f | M = 0 } M ⊥ = f ∈ Li p o [ a , b ] ; f M = 0 M^(_|_)={f in Lip_(o)[a,b];f|_(M)=0}M^{\perp}=\left\{f \in \operatorname{Li} p_{o}[a, b] ;\left.f\right|_{M}=0\right\}M⊥={f∈Lipo[a,b];f|M=0}
Since every function f ∈ Lip 0 M f ∈ Lip 0 M f inLip_(0)Mf \in \operatorname{Lip}_{0} Mf∈Lip0M has at least one extension F ∈ Lip 0 [ a , b ] F ∈ Lip 0 [ a , b ] F inLip_(0)[a,b]F \in \operatorname{Lip}_{0}[a, b]F∈Lip0[a,b] it follows that every function f ∈ Lip 0 [ a , b ] f ∈ Lip 0 [ a , b ] f inLip_(0)[a,b]f \in \operatorname{Lip}_{0}[a, b]f∈Lip0[a,b] has a best approximation (nearest point) in M ⊥ M ⊥ M^(_|_)M^{\perp}M⊥, i.e. there exists E 0 ∈ M ⊥ E 0 ∈ M ⊥ E_(0)inM^(_|_)\mathrm{E}_{0} \in \mathbb{M}^{\perp}E0∈M⊥. such that ‖ f − g 0 ‖ L = inf { ‖ P − g ‖ L : g ∈ M ⊥ } f − g 0 L = inf ‖ P − g ‖ L : g ∈ M ⊥ ||f-g_(0)||_(L)=i n f{||P-g||_(L):g inM^(_|_)}\left\|\mathrm{f}-\mathrm{g}_{0}\right\|_{\mathrm{L}}=\inf \left\{\|\mathrm{P}-\mathrm{g}\|_{\mathrm{L}}: g \in \mathbb{M}^{\perp}\right\}‖f−g0‖L=inf{‖P−g‖L:g∈M⊥}. It Was shown ( see [5] ) that g 0 ∈ M ⊥ g 0 ∈ M ⊥ g_(0)inM^(_|_)g_{0} \in M^{\perp}g0∈M⊥ is an element of best approximation for f ∈ Lip 0 [ a , b ] f ∈ Lip 0 [ a , b ] f inLip_(0)[a,b]f \in \operatorname{Lip}_{0}[a, b]f∈Lip0[a,b] by elements from M ⊥ M ⊥ M^(_|_)M^{\perp}M⊥ if and only if C 0 = f − F C 0 = f − F C_(0)=f-FC_{0}=f-FC0=f−F, for F ∈ V ( f ) F ∈ V ( f ) F in V(f)F \in V(f)F∈V(f).
Taking into account the precedings results it follows :
4.1 THEOREIM. Let f ∈ Lip p 0 [ a , b ] f ∈ Lip p 0 [ a , b ] f in Lipp_(0)[a,b]f \in \operatorname{Lip} p_{0}[a, b]f∈Lipp0[a,b] and M = { x 0 , x 1 , … , x n } M = x 0 , x 1 , … , x n M={x_(0),x_(1),dots,x_(n)}M=\left\{x_{0}, x_{1}, \ldots, x_{n}\right\}M={x0,x1,…,xn}, a ⩽ x 0 < x 1 < … < x n ⩽ b a ⩽ x 0 < x 1 < … < x n ⩽ b a <= x_(0) < x_(1) < dots < x_(n) <= ba \leqslant x_{0}<x_{1}<\ldots<x_{n} \leqslant ba⩽x0<x1<…<xn⩽b.
(a) If the relation (15) holds for an index k ∈ { 0 , 1 , … , n − 1 } k ∈ { 0 , 1 , … , n − 1 } k in{0,1,dots,n-1}k \in\{0,1, \ldots, n-1\}k∈{0,1,…,n−1} then all the best approximation elements for f f fff in M ⊥ M ⊥ M^(_|_)M^{\perp}M⊥ vanish on the interval [ x k , x k + 1 ] x k , x k + 1 [x_(k),x_(k+1)]\left[\mathrm{x}_{\mathrm{k}}, \mathrm{x}_{\mathrm{k}+1}\right][xk,xk+1];
(b) If x 0 = a , x n = b x 0 = a , x n = b x_(0)=a,x_(n)=bx_{0}=a, x_{n}=bx0=a,xn=b and ‖ f ‖ L = | [ x 0 , x n ; f ] | ‖ f ‖ L = x 0 , x n ; f ||f||_(L)=|[x_(0),x_(n);f]|\|f\|_{L}=\left|\left[x_{0}, x_{n} ; f\right]\right|‖f‖L=|[x0,xn;f]|, then 0 is the only best approximation element for f f fff in M M MMM;
(c) If ‖ f ‖ L = | [ x k , x k + 1 ; f ] | ‖ f ‖ L = x k , x k + 1 ; f ||f||_(L)=|[x_(k),x_(k+1);f]|\|f\|_{L}=\left|\left[x_{k}, x_{k+1} ; f\right]\right|‖f‖L=|[xk,xk+1;f]|, for all k = 0 , 1 , … , n − 1 k = 0 , 1 , … , n − 1 k=0,1,dots,n-1k=0,1, \ldots, n-1k=0,1,…,n−1, then 0 is the only best approximation olement for it in u ⊥ u ⊥ u^(_|_)u^{\perp}u⊥.
Proof. Assertion (a) follows from Corollary 2.3 , taking into account that the best approximation elements g 0 g 0 g_(0)g_{0}g0 for f f fff in ii ⊥ ⊥ ^(_|_){ }^{\perp}⊥ have the form
g 0 = I − E , F ∈ V ( P ) g 0 = I − E , F ∈ V ( P ) g_(0)=I-E quad,quad F inV(P)g_{0}=I-E \quad, \quad F \in \mathbb{V}(P)g0=I−E,F∈V(P)
Assertions (b) and (c) follow from Corollary 2.4 .
4.2 Remark. Let M M MMM be the set (8) and let
Lip 0 [ a , b ] = { f : [ a , b ] → R , f ( x 0 ) = 0 , f is Lipschitz } Lip 0 [ a , b ] = f : [ a , b ] → R , f x 0 = 0 , f  is Lipschitz  Lip_(0)[a,b]={f:[a,b]rarr R,f(x_(0))=0,f" is Lipschitz "}\operatorname{Lip}_{0}[a, b]=\left\{f:[a, b] \rightarrow R, f\left(x_{0}\right)=0, f \text { is Lipschitz }\right\}Lip0[a,b]={f:[a,b]→R,f(x0)=0,f is Lipschitz }
If f ∉ W ( f ) f ∉ W ( f ) f!in W(f)f \notin W(f)f∉W(f), then, by Theorem 2.1, it follws
‖ f ‖ L > ‖ f | M ‖ L ‖ f ‖ L > f M L ||f||_(L) > ||f|_(M)||_(L)\|f\|_{L}>\left\|\left.f\right|_{M}\right\|_{L}‖f‖L>‖f|M‖L
If, further
(i)
f ( x ) ⩽ F 1 ( x ) , x ∈ [ a , b ] f ( x ) ⩽ F 1 ( x ) , x ∈ [ a , b ] f(x) <= F_(1)(x)quad,quad x in[a,b]f(x) \leqslant F_{1}(x) \quad, \quad x \in[a, b]f(x)⩽F1(x),x∈[a,b]
where F 1 F 1 F_(1)F_{1}F1 is given by (5) (with A = M A = M A=MA=MA=M ), then all bect approximation elements of f f fff in M ⊥ M ⊥ M^(_|_)M^{\perp}M⊥ are non-positive, and if
(ii) f ( x ) ⩾ F 2 ( x ) , x ∈ [ a , b ] (ii) f ( x ) ⩾ F 2 ( x ) , x ∈ [ a , b ] {:(ii)f(x) >= F_(2)(x)quad","quad x in[a","b]:}\begin{equation*} f(x) \geqslant F_{2}(x) \quad, \quad x \in[a, b] \tag{ii} \end{equation*}(ii)f(x)⩾F2(x),x∈[a,b]
with F 2 F 2 F_(2)F_{2}F2 given by (6) (with A = M A = M A=MA=MA=M ), then all the elements of best approximation for f f fff in M ⊥ M ⊥ M^(_|_)M^{\perp}M⊥ are non-negative.
Obviously, there exist functions f ∈ Lip o [ a , b ] f ∈ Lip o [ a , b ] f inLip_(o)[a,b]f \in \operatorname{Lip}_{o}[a, b]f∈Lipo[a,b] verifying the conditions (i) and (ii) and which are not fixed points of W W WWW.
For example, takin G g ∈ Lip o [ a , b ] G g ∈ Lip o [ a , b ] _(G)g inLip_(o)[a,b]{ }_{G} g \in \operatorname{Lip}_{o}[a, b]Gg∈Lipo[a,b], the function
f ( x ) = sup { g ( x k ) − L i | x − x k | : x k ∈ M , k = 0 , 1 , … , n } f ( x ) = sup g x k − L i x − x k : x k ∈ M , k = 0 , 1 , … , n f(x)=s u p{g(x_(k))-L_(i)|x-x_(k)|:x_(k)in M,k=0,1,dots,n}f(x)=\sup \left\{g\left(x_{k}\right)-L_{i}\left|x-x_{k}\right|: x_{k} \in M, k=0,1, \ldots, n\right\}f(x)=sup{g(xk)−Li|x−xk|:xk∈M,k=0,1,…,n}
with L > ‖ g | M ‖ L L > g M L L > ||g|_(M)||_(L)L>\left\|\left.g\right|_{\mathbb{M}}\right\|_{L}L>‖g|M‖L, verifies the condition (i) and the function
h ( x ) = inf { g ( x k ) + I . | x − x k | : x k ∈ h , k = 0 , 1 , … , n } h ( x ) = inf g x k + I . x − x k : x k ∈ h , k = 0 , 1 , … , n h(x)=i n f{g(x_(k))+I.|x-x_(k)|:x_(k)in h,k=0,1,dots,n}h(x)=\inf \left\{g\left(x_{k}\right)+I .\left|x-x_{k}\right|: x_{k} \in h, k=0,1, \ldots, n\right\}h(x)=inf{g(xk)+I.|x−xk|:xk∈h,k=0,1,…,n}
with L > ‖ B | ‖ ‖ I L > B ‖ I L > ||B|_(||)||_(I)L>\left\|\left.B\right|_{\|}\right\|_{I}L>‖B|‖‖I verifies the condition h ( x ) ⩽ I 2 ( x ) h ( x ) ⩽ I 2 ( x ) h(x) <= I_(2)(x)h(x) \leqslant I_{2}(x)h(x)⩽I2(x), x ∈ [ a , b ] x ∈ [ a , b ] x in[a,b]x \in[a, b]x∈[a,b] (i.e. condition (ii) ) .
1 COBZAS, S. , MUSTATA, C. , Norm Preserving Extension of Convex Iipschitz Functions, J.A.T. 24 (1978), 236 - 244.
2 DUTHAS, O.B. , Ohebyshev approximation with a null space , Proc. Amer. Math. Soc. 41 (1973), 557-558.
3 JOLINSON, J. A., Banach Spaces of Lipschitz Functions and Vector-Valued Lipschitz Functions, Trans. Amer. Math. Soc. 148 (1970), 147-169.
4 Wic SHATL, E.J., Extension of range of functions, Bull. Amer. Jath. Soc. 40 (1934), 837-842.
5 USTATA, C., Best Approximation and Unique Intiension of Lipsch itz Functions , J.A.T. 19 (1977), 222-230.
6 R.OI, A.I., Extreme Points and Iinear Isometries of Banach Space of Lijschitz Functions, Canad. J. of Math. 20 (1968), 1150-1164.
7 SIMGR, I., Cea nai bună aproximare in spatii vectoriale normate prin elemente din subspatii vectoriale, Edit.
Acad. R.S. România, Bucureşti, 1967 .

Related Posts