Principe de délimitation uniforme

Principe de délimitation uniforme (Redirected from Banach–Steinhaus theorem) Jump to navigation Jump to search For the definition of uniformly bounded functions, see Uniform boundedness. For the conjectures in number theory and algebraic geometry, see Uniform boundedness conjecture.

En mathématiques, the uniform boundedness principle or Banach–Steinhaus theorem is one of the fundamental results in functional analysis. Together with the Hahn–Banach theorem and the open mapping theorem, it is considered one of the cornerstones of the field. In its basic form, it asserts that for a family of continuous linear operators (and thus bounded operators) whose domain is a Banach space, pointwise boundedness is equivalent to uniform boundedness in operator norm.

The theorem was first published in 1927 by Stefan Banach and Hugo Steinhaus, but it was also proven independently by Hans Hahn.

Contenu 1 Théorème 2 Corollaires 3 Exemple: pointwise convergence of Fourier series 4 Généralisations 4.1 Barrelled spaces 4.2 Uniform boundedness in topological vector spaces 4.3 Generalizations involving nonmeager subsets 4.3.1 Complete metrizable domain 5 Voir également 6 Remarques 7 Citations 8 Bibliography Theorem Uniform Boundedness Principle — Let {style d'affichage X} be a Banach space, {style d'affichage Y} a normed vector space and {style d'affichage B(X,Oui)} the space of all continuous linear operators from {style d'affichage X} dans {style d'affichage Y} . Supposer que {style d'affichage F} is a collection of continuous linear operators from {style d'affichage X} à {style d'affichage Y.} Si {displaystyle sup _{Tin F}|J(X)|_{Oui}0} tel que {style d'affichage {surligner {B_{varepsilon }(X_{0})}}~:=~left{xin X,:,|x-x_{0}|leq varepsilon right}~subseteq ~X_{m}.} Laisser {displaystyle uin X} avec {style d'affichage |tu|leq 1} et {displaystyle Tin F.} Alors: {style d'affichage {commencer{aligné}|J(tu)|_{Oui}&=varepsilon ^{-1}la gauche|Tleft(X_{0}+varepsilon uright)-Tleft(X_{0}droit)droit|_{Oui}&[{texte{ by linearity of }}J]\&leq varepsilon ^{-1}la gauche(la gauche|J(X_{0}+varepsilon u)droit|_{Oui}+la gauche|J(X_{0})droit|_{Oui}droit)\&leq varepsilon ^{-1}(m+m).&[{texte{ puisque }} X_{0}+varepsilon u, X_{0}in X_{m}]\fin{aligné}}} Taking the supremum over {style d'affichage u} in the unit ball of {style d'affichage X} and over {displaystyle Tin F} il s'ensuit que {displaystyle sup _{Tin F}|J|_{B(X,Oui)}~leq ~2varepsilon ^{-1}m~<~infty .} There are also simple proofs not using the Baire theorem (Sokal 2011). Corollaries Corollary — If a sequence of bounded operators {displaystyle left(T_{n}right)} converges pointwise, that is, the limit of {displaystyle left(T_{n}(x)right)} exists for all {displaystyle xin X,} then these pointwise limits define a bounded linear operator {displaystyle T.} The above corollary does not claim that {displaystyle T_{n}} converges to {displaystyle T} in operator norm, that is, uniformly on bounded sets. However, since {displaystyle left{T_{n}right}} is bounded in operator norm, and the limit operator {displaystyle T} is continuous, a standard " {displaystyle 3varepsilon } " estimate shows that {displaystyle T_{n}} converges to {displaystyle T} uniformly on compact sets. Proof Essentially the same as that of the proof that a pointwise convergent sequence of uniformly continuous functions on a compact set converges to a continuous function. By uniform boundedness principle, let {displaystyle M=max{sup _{n}|T_{n}|,T}} be a uniform upper bound on the operator norms. Fix any compact {displaystyle Ksubset X} . Then for any {displaystyle epsilon >0} , finitely cover (use compactness) {style d'affichage K} by a finite set of open balls {style d'affichage {B(X_{je},r)}_{i=1,...,N}} of radius {displaystyle r={frac {epsilon }{M}}} Depuis {style d'affichage T_{n}to T} pointwise on each of {style d'affichage x_{1},...,X_{N}} , for all large {displaystyle n} , {style d'affichage |T_{n}(X_{je})-J(X_{je})|leq epsilon } pour tous {displaystyle i=1,...,N} .

Then by triangle inequality, we find for all large {displaystyle n} , {displaystyle forall xin K,|T_{n}(X_{je})-J(X_{je})|leq 3epsilon } .

Corollary — Any weakly bounded subset {displaystyle Ssubseteq Y} in a normed space {style d'affichage Y} est délimité.

En effet, les éléments de {style d'affichage S} define a pointwise bounded family of continuous linear forms on the Banach space {style d'affichage X:=Y',} which is the continuous dual space of {style d'affichage Y.} By the uniform boundedness principle, the norms of elements of {style d'affichage S,} as functionals on {style d'affichage X,} C'est, norms in the second dual {displaystyle Y'',} are bounded. But for every {displaystyle sin S,} the norm in the second dual coincides with the norm in {style d'affichage Y,} by a consequence of the Hahn–Banach theorem.

Laisser {displaystyle L(X,Oui)} denote the continuous operators from {style d'affichage X} à {style d'affichage Y,} endowed with the operator norm. If the collection {style d'affichage F} is unbounded in {displaystyle L(X,Oui),} then the uniform boundedness principle implies: {displaystyle R=left{xin X : sup nolimits _{Tin F}|Tx|_{Oui}=infty right}neq varnothing .} En réalité, {style d'affichage R} est dense en {style d'affichage X.} The complement of {style d'affichage R} dans {style d'affichage X} is the countable union of closed sets {textstyle bigcup X_{n}.} By the argument used in proving the theorem, each {style d'affichage X_{n}} is nowhere dense, c'est à dire. le sous-ensemble {textstyle bigcup X_{n}} is of first category. Par conséquent {style d'affichage R} is the complement of a subset of first category in a Baire space. By definition of a Baire space, such sets (called comeagre or residual sets) are dense. Such reasoning leads to the principle of condensation of singularities, which can be formulated as follows: Theorem — Let {style d'affichage X} be a Banach space, {style d'affichage à gauche(O_{n}droit)} a sequence of normed vector spaces, and for every {displaystyle n,} laisser {style d'affichage F_{n}} an unbounded family in {displaystyle Lleft(X,O_{n}droit).} Then the set {style d'affichage R:=gauche{xin X : {texte{ pour tous }}nin mathbb {N} ,souper _{Tin F_{n}}|Tx|_{O_{n}}=infty right}} is a residual set, and thus dense in {style d'affichage X.} Proof The complement of {style d'affichage R} is the countable union {style d'affichage bigcup _{n,m}la gauche{xin X : souper _{Tin F_{n}}|Tx|_{O_{n}}leq mright}} of sets of first category. Par conséquent, its residual set {style d'affichage R} is dense.

Exemple: pointwise convergence of Fourier series Let {style d'affichage mathbb {J} } be the circle, et laissez {displaystyle C(mathbb {J} )} be the Banach space of continuous functions on {style d'affichage mathbb {J} ,} with the uniform norm. Using the uniform boundedness principle, one can show that there exists an element in {displaystyle C(mathbb {J} )} for which the Fourier series does not converge pointwise.

Pour {displaystyle fin C(mathbb {J} ),} its Fourier series is defined by {somme de style d'affichage _{kin mathbb {Z} }{chapeau {F}}(k)e ^{ikx}=somme _{kin mathbb {Z} }{frac {1}{2pi }}la gauche(entier _{0}^{2pi }F(t)e ^{-ikt}c'est vrai)e ^{ikx},} and the N-th symmetric partial sum is {style d'affichage S_{N}(F)(X)=somme _{k=-N}^{N}{chapeau {F}}(k)e ^{ikx}={frac {1}{2pi }}entier _{0}^{2pi }F(t)RÉ_{N}(x-t),dt,} où {displaystyle D_{N}} est le {displaystyle N} -th Dirichlet kernel. Réparer {style d'affichage xin mathbb {J} } and consider the convergence of {style d'affichage à gauche{S_{N}(F)(X)droit}.} The functional {style d'affichage varphi _{N,X}:C(mathbb {J} )à mathbb {C} } Défini par {style d'affichage varphi _{N,X}(F)=S_{N}(F)(X),qquad fin C(mathbb {J} ),} est délimité. The norm of {style d'affichage varphi _{N,X},} in the dual of {displaystyle C(mathbb {J} ),} is the norm of the signed measure {style d'affichage (2(2pi )^{-1}RÉ_{N}(x-t)dt,} à savoir {style d'affichage à gauche|varphi _{N,X}droit|={frac {1}{2pi }}entier _{0}^{2pi }la gauche|RÉ_{N}(x-t)droit|,dt={frac {1}{2pi }}entier _{0}^{2pi }la gauche|RÉ_{N}(s)droit|,ds=left|RÉ_{N}droit|_{L^{1}(mathbb {J} )}.} It can be verified that {style d'affichage {frac {1}{2pi }}entier _{0}^{2pi }|RÉ_{N}(t)|,dtgeq {frac {1}{2pi }}entier _{0}^{2pi }{frac {la gauche|sin left((N+{tfrac {1}{2}})tright)droit|}{t/2}},dtto infty .} So the collection {style d'affichage à gauche(varphi _{N,X}droit)} is unbounded in {displaystyle C(mathbb {J} )^{dernièrement },} le duel de {displaystyle C(mathbb {J} ).} Par conséquent, by the uniform boundedness principle, pour toute {style d'affichage xin mathbb {J} ,} the set of continuous functions whose Fourier series diverges at {style d'affichage x} est dense en {displaystyle C(mathbb {J} ).} More can be concluded by applying the principle of condensation of singularities. Laisser {style d'affichage à gauche(X_{m}droit)} be a dense sequence in {style d'affichage mathbb {J} .} Définir {style d'affichage varphi _{N,X_{m}}} in the similar way as above. The principle of condensation of singularities then says that the set of continuous functions whose Fourier series diverges at each {style d'affichage x_{m}} est dense en {displaystyle C(mathbb {J} )} (toutefois, the Fourier series of a continuous function {style d'affichage f} converge vers {style d'affichage f(X)} for almost every {style d'affichage xin mathbb {J} ,} by Carleson's theorem).

Generalizations In a topological vector space (téléviseurs) {style d'affichage X,} "bounded subset" refers specifically to the notion of a von Neumann bounded subset. Si {style d'affichage X} happens to also be a normed or seminormed space, say with (semi)norme {style d'affichage |cdot |,} then a subset {style d'affichage B} est (par Neumann) bounded if and only if it is norm bounded, which by definition means {textstyle sup _{bin B}|b|0} {displaystyle tgeq r} {displaystyle Csubseteq tU.} {displaystyle tgeq r,} {displaystyle h(C)subseteq h(tU)=th(U)subseteq tV,} {textstyle bigcup _{hin H}h(C)subseteq tV.} Citations Shtern 2001. Jump up to: b c d Rudin 1991, pp. 42−47. Bibliography Banach, Stefan; Steinhaus, Hugo (1927), "Sur le principe de la condensation singularités" (PDF), Fundamenta Mathematicae, 9: 50–61, doi:10.4064>

Si vous voulez connaître d'autres articles similaires à Principe de délimitation uniforme vous pouvez visiter la catégorie Analyse fonctionnelle.

Laisser un commentaire

Votre adresse email ne sera pas publiée.


Nous utilisons nos propres cookies et ceux de tiers pour améliorer l'expérience utilisateur Plus d'informations