Hahn–Banach theorem

Template:Short description In functional analysis, the Hahn–Banach theorem is a central result that allows the extension of bounded linear functionals defined on a vector subspace of some vector space to the whole space. The theorem also shows that there are sufficient continuous linear functionals defined on every normed vector space in order to study the dual space. Another version of the Hahn–Banach theorem is known as the Hahn–Banach separation theorem or the hyperplane separation theorem, and has numerous uses in convex geometry.

History

The theorem is named for the mathematicians Hans Hahn and Stefan Banach, who proved it independently in the late 1920s. The special case of the theorem for the space $C[a,b]$ of continuous functions on an interval was proved earlier (in 1912) by Eduard Helly, and a more general extension theorem, the M. Riesz extension theorem, from which the Hahn–Banach theorem can be derived, was proved in 1923 by Marcel Riesz.

The first Hahn–Banach theorem was proved by Eduard Helly in 1912 who showed that certain linear functionals defined on a subspace of a certain type of normed space (${ℂ}^{\mathbb{N}}$) had an extension of the same norm. Helly did this through the technique of first proving that a one-dimensional extension exists (where the linear functional has its domain extended by one dimension) and then using induction. In 1927, Hahn defined general Banach spaces and used Helly's technique to prove a norm-preserving version of Hahn–Banach theorem for Banach spaces (where a bounded linear functional on a subspace has a bounded linear extension of the same norm to the whole space). In 1929, Banach, who was unaware of Hahn's result, generalized it by replacing the norm-preserving version with the dominated extension version that uses sublinear functions. Whereas Helly's proof used mathematical induction, Hahn and Banach both used transfinite induction.Template:Sfn

The Hahn–Banach theorem arose from attempts to solve infinite systems of linear equations. This is needed to solve problems such as the moment problem, whereby given all the potential moments of a function one must determine if a function having these moments exists, and, if so, find it in terms of those moments. Another such problem is the Fourier cosine series problem, whereby given all the potential Fourier cosine coefficients one must determine if a function having those coefficients exists, and, again, find it if so.

Riesz and Helly solved the problem for certain classes of spaces (such as $L^p([0,1])$ and $C([a,b])$) where they discovered that the existence of a solution was equivalent to the existence and continuity of certain linear functionals. In effect, they needed to solve the following problem:Template:Sfn

(Template:Visible anchor) Given a collection ${\left(f_i\right)}_{i\in I}$ of bounded linear functionals on a normed space $X$ and a collection of scalars ${\left(c_i\right)}_{i\in I},$ determine if there is an $x\in X$ such that $f_i(x)=c_i$ for all $i\in I.$

If $X$ happens to be a reflexive space then to solve the vector problem, it suffices to solve the following dual problem:Template:Sfn

(The functional problem) Given a collection ${\left(x_i\right)}_{i\in I}$ of vectors in a normed space $X$ and a collection of scalars ${\left(c_i\right)}_{i\in I},$ determine if there is a bounded linear functional $f$ on $X$ such that $f\left(x_i\right)=c_i$ for all $i\in I.$

Riesz went on to define $L^p([0,1])$ space ($1<p<\mathrm{\infty }$) in 1910 and the ${\ell }^p$ spaces in 1913. While investigating these spaces he proved a special case of the Hahn–Banach theorem. Helly also proved a special case of the Hahn–Banach theorem in 1912. In 1910, Riesz solved the functional problem for some specific spaces and in 1912, Helly solved it for a more general class of spaces. It wasn't until 1932 that Banach, in one of the first important applications of the Hahn–Banach theorem, solved the general functional problem. The following theorem states the general functional problem and characterizes its solution.Template:Sfn

The Hahn–Banach theorem can be deduced from the above theorem.Template:Sfn If $X$ is reflexive then this theorem solves the vector problem.

Hahn–Banach theorem

A real-valued function $f:M\to ℝ$ defined on a subset $M$ of $X$ is said to be Template:Em a function $p:X\to ℝ$ if $f(m)\le p(m)$ for every $m\in M.$ For this reason, the following version of the Hahn–Banach theorem is called Template:Em.

The theorem remains true if the requirements on $p$ are relaxed to require only that $p$ be a convex function:Template:SfnTemplate:Sfn $$p(tx+(1-t)y)\le tp(x)+(1-t)p(y)\text{ for all }0<t<1\text{ and }x,y\in X.$$ A function $p:X\to ℝ$ is convex and satisfies $p(0)\le 0$ if and only if $p(ax+by)\le ap(x)+bp(y)$ for all vectors $x,y\in X$ and all non-negative real $a,b\ge 0$ such that $a+b\le 1.$ Every sublinear function is a convex function. On the other hand, if $p:X\to ℝ$ is convex with $p(0)\ge 0,$ then the function defined by $p_0(x)\stackrel{\text{def}}{=}\underset{t>0}{\inf }\frac{p(tx)}{t}$ is positively homogeneous (because for all $x$ and $r>0$ one has $p_0(rx)=\underset{t>0}{\inf }\frac{p(trx)}{t}=r\underset{t>0}{\inf }\frac{p(trx)}{tr}=r\underset{\tau >0}{\inf }\frac{p(\tau x)}{\tau }=r{p}_0(x)$), hence, being convex, it is sublinear. It is also bounded above by $p_0\le p,$ and satisfies $F\le p_0$ for every linear functional $F\le p.$ So the extension of the Hahn–Banach theorem to convex functionals does not have a much larger content than the classical one stated for sublinear functionals.

If $F:X\to ℝ$ is linear then $F\le p$ if and only ifTemplate:Sfn $$-p(-x)\le F(x)\le p(x)\text{ for all }x\in X,$$ which is the (equivalent) conclusion that some authorsTemplate:Sfn write instead of $F\le p.$ It follows that if $p:X\to ℝ$ is also Template:Em, meaning that $p(-x)=p(x)$ holds for all $x\in X,$ then $F\le p$ if and only $|F|\le p.$ Every norm is a seminorm and both are symmetric balanced sublinear functions. A sublinear function is a seminorm if and only if it is a balanced function. On a real vector space (although not on a complex vector space), a sublinear function is a seminorm if and only if it is symmetric. The identity function $ℝ\to ℝ$ on $X:=ℝ$ is an example of a sublinear function that is not a seminorm.

For complex or real vector spaces

The dominated extension theorem for real linear functionals implies the following alternative statement of the Hahn–Banach theorem that can be applied to linear functionals on real or complex vector spaces.

The theorem remains true if the requirements on $p$ are relaxed to require only that for all $x,y\in X$ and all scalars $a$ and $b$ satisfying $|a|+|b|\le 1,$Template:Sfn $$p(ax+by)\le |a|p(x)+|b|p(y).$$ This condition holds if and only if $p$ is a convex and balanced function satisfying $p(0)\le 0,$ or equivalently, if and only if it is convex, satisfies $p(0)\le 0,$ and $p(ux)\le p(x)$ for all $x\in X$ and all unit length scalars $u.$

A complex-valued functional $F$ is said to be Template:Em if $|F(x)|\le p(x)$ for all $x$ in the domain of $F.$ With this terminology, the above statements of the Hahn–Banach theorem can be restated more succinctly:

Hahn–Banach dominated extension theorem: If $p:X\to ℝ$ is a seminorm defined on a real or complex vector space $X,$ then every dominated linear functional defined on a vector subspace of $X$ has a dominated linear extension to all of $X.$ In the case where $X$ is a real vector space and $p:X\to ℝ$ is merely a convex or sublinear function, this conclusion will remain true if both instances of "dominated" (meaning $|F|\le p$) are weakened to instead mean "[[#dominated real functional|dominated Template:Em]]" (meaning $F\le p$).Template:SfnTemplate:Sfn

Proof

The following observations allow the Hahn–Banach theorem for real vector spaces to be applied to (complex-valued) linear functionals on complex vector spaces.

Every linear functional $F:X\to ℂ$ on a complex vector space is completely determined by its real part $\mathrm{Re}F:X\to ℝ$ through the formulaTemplate:Sfn $$F(x)=\mathrm{Re}F(x)-i\mathrm{Re}F(ix)\text{ for all }x\in X$$ and moreover, if $\Vert \cdot \Vert$ is a norm on $X$ then their dual norms are equal: $\Vert F\Vert =\Vert \mathrm{Re}F\Vert .$Template:Sfn In particular, a linear functional on $X$ extends another one defined on $M\subseteq X$ if and only if their real parts are equal on $M$ (in other words, a linear functional $F$ extends $f$ if and only if $\mathrm{Re}F$ extends $\mathrm{Re}f$). The real part of a linear functional on $X$ is always a Template:Visible anchor (meaning that it is linear when $X$ is considered as a real vector space) and if $R:X\to ℝ$ is a real-linear functional on a complex vector space then $x\mapsto R(x)-iR(ix)$ defines the unique linear functional on $X$ whose real part is $R.$

If $F$ is a linear functional on a (complex or real) vector space $X$ and if $p:X\to ℝ$ is a seminorm thenTemplate:Sfn $$|F|\le p\text{ if and only if }\mathrm{Re}F\le p.$$ Stated in simpler language, a linear functional is dominated by a seminorm $p$ if and only if its real part is dominated above by $p.$

The proof above shows that when $p$ is a seminorm then there is a one-to-one correspondence between dominated linear extensions of $f:M\to ℂ$ and dominated real-linear extensions of $\mathrm{Re}f:M\to ℝ;$ the proof even gives a formula for explicitly constructing a linear extension of $f$ from any given real-linear extension of its real part.

Continuity

A linear functional $F$ on a topological vector space is continuous if and only if this is true of its real part $\mathrm{Re}F;$ if the domain is a normed space then $\Vert F\Vert =\Vert \mathrm{Re}F\Vert$ (where one side is infinite if and only if the other side is infinite).Template:Sfn Assume $X$ is a topological vector space and $p:X\to ℝ$ is sublinear function. If $p$ is a continuous sublinear function that dominates a linear functional $F$ then $F$ is necessarily continuous.Template:Sfn Moreover, a linear functional $F$ is continuous if and only if its absolute value $|F|$ (which is a seminorm that dominates $F$) is continuous.Template:Sfn In particular, a linear functional is continuous if and only if it is dominated by some continuous sublinear function.

Proof

The Hahn–Banach theorem for real vector spaces ultimately follows from Helly's initial result for the special case where the linear functional is extended from $M$ to a larger vector space in which $M$ has codimension $1.$Template:Sfn

This lemma remains true if $p:X\to ℝ$ is merely a convex function instead of a sublinear function.Template:SfnTemplate:Sfn

Template:Collapse top Assume that $p$ is convex, which means that $p(ty+(1-t)z)\le tp(y)+(1-t)p(z)$ for all $0\le t\le 1$ and $y,z\in X.$ Let $M,$ $f:M\to ℝ,$ and $x\in X\setminus M$ be as in the lemma's statement. Given any $m,n\in M$ and any positive real $r,s>0,$ the positive real numbers $t:=\frac{s}{r+s}$ and $\frac{r}{r+s}=1-t$ sum to $1$ so that the convexity of $p$ on $X$ guarantees $$\begin{array}{cccccc}p(\frac{s}{r+s}m+\frac{r}{r+s}n)& =p(\frac{s}{r+s}(m-rx)& & +\frac{r}{r+s}(n+sx))& & \\ & \le \frac{s}{r+s}p(m-rx)& & +\frac{r}{r+s}p(n+sx)& & \\ \end{array}$$ and hence $$\begin{array}{cccc}sf(m)+rf(n)& =(r+s)f(\frac{s}{r+s}m+\frac{r}{r+s}n)& & \text{ by linearity of }f\\ & \le (r+s)p(\frac{s}{r+s}m+\frac{r}{r+s}n)& & f\le p\text{ on }M\\ & \le sp(m-rx)+rp(n+sx)\\ \end{array}$$ thus proving that $-sp(m-rx)+sf(m)\le rp(n+sx)-rf(n),$ which after multiplying both sides by $\frac{1}{rs}$ becomes $$\frac{1}{r}[-p(m-rx)+f(m)]\le \frac{1}{s}[p(n+sx)-f(n)].$$ This implies that the values defined by $$a=\underset{\stackrel{m\in M}{r>0}}{\sup }\frac{1}{r}[-p(m-rx)+f(m)]\text{ and }c=\underset{\stackrel{n\in M}{s>0}}{\inf }\frac{1}{s}[p(n+sx)-f(n)]$$ are real numbers that satisfy $a\le c.$ As in the above proof of the one–dimensional dominated extension theorem above, for any real $b\in ℝ$ define $F_b:M\oplus ℝx\to ℝ$ by $F_b(m+rx)=f(m)+rb.$ It can be verified that if $a\le b\le c$ then $F_b\le p$ where $rb\le p(m+rx)-f(m)$ follows from $b\le c$ when $r>0$ (respectively, follows from $a\le b$ when $r<0$). $◼$ Template:Collapse bottom

The lemma above is the key step in deducing the dominated extension theorem from Zorn's lemma.

When $M$ has countable codimension, then using induction and the lemma completes the proof of the Hahn–Banach theorem. The standard proof of the general case uses Zorn's lemma although the strictly weaker ultrafilter lemmaTemplate:Sfn (which is equivalent to the compactness theorem and to the Boolean prime ideal theorem) may be used instead. Hahn–Banach can also be proved using Tychonoff's theorem for compact Hausdorff spacesTemplate:Sfn (which is also equivalent to the ultrafilter lemma)

The Mizar project has completely formalized and automatically checked the proof of the Hahn–Banach theorem in the HAHNBAN file.

Continuous extension theorem

The Hahn–Banach theorem can be used to guarantee the existence of continuous linear extensions of continuous linear functionals.

In category-theoretic terms, the underlying field of the vector space is an injective object in the category of locally convex vector spaces.

On a normed (or seminormed) space, a linear extension $F$ of a bounded linear functional $f$ is said to be Template:Em if it has the same dual norm as the original functional: $\Vert F\Vert =\Vert f\Vert .$ Because of this terminology, the second part of the above theorem is sometimes referred to as the "norm-preserving" version of the Hahn–Banach theorem.Template:Sfn Explicitly:

Proof of the continuous extension theorem

The following observations allow the continuous extension theorem to be deduced from the Hahn–Banach theorem.Template:Sfn

The absolute value of a linear functional is always a seminorm. A linear functional $F$ on a topological vector space $X$ is continuous if and only if its absolute value $|F|$ is continuous, which happens if and only if there exists a continuous seminorm $p$ on $X$ such that $|F|\le p$ on the domain of $F.$Template:Sfn If $X$ is a locally convex space then this statement remains true when the linear functional $F$ is defined on a Template:Em vector subspace of $X.$

Proof for normed spaces

A linear functional $f$ on a normed space is continuous if and only if it is bounded, which means that its dual norm $$\Vert f\Vert =\sup \{|f(m)|:\Vert m\Vert \le 1,m\in \mathrm{domain}f\}$$ is finite, in which case $|f(m)|\le \Vert f\Vert \Vert m\Vert$ holds for every point $m$ in its domain. Moreover, if $c\ge 0$ is such that $|f(m)|\le c\Vert m\Vert$ for all $m$ in the functional's domain, then necessarily $\Vert f\Vert \le c.$ If $F$ is a linear extension of a linear functional $f$ then their dual norms always satisfy $\Vert f\Vert \le \Vert F\Vert$ so that equality $\Vert f\Vert =\Vert F\Vert$ is equivalent to $\Vert F\Vert \le \Vert f\Vert ,$ which holds if and only if $|F(x)|\le \Vert f\Vert \Vert x\Vert$ for every point $x$ in the extension's domain. This can be restated in terms of the function $\Vert f\Vert \Vert \cdot \Vert :X\to ℝ$ defined by $x\mapsto \Vert f\Vert \Vert x\Vert ,$ which is always a seminorm:

A linear extension of a bounded linear functional $f$ is norm-preserving if and only if the extension is dominated by the seminorm $\Vert f\Vert \Vert \cdot \Vert .$

Applying the Hahn–Banach theorem to $f$ with this seminorm $\Vert f\Vert \Vert \cdot \Vert$ thus produces a dominated linear extension whose norm is (necessarily) equal to that of $f,$ which proves the theorem:

Non-locally convex spaces

The continuous extension theorem might fail if the topological vector space (TVS) $X$ is not locally convex. For example, for $0<p<1,$ the Lebesgue space $L^p([0,1])$ is a complete metrizable TVS (an F-space) that is Template:Em locally convex (in fact, its only convex open subsets are itself $L^p([0,1])$ and the empty set) and the only continuous linear functional on $L^p([0,1])$ is the constant $0$ function Template:Harv. Since $L^p([0,1])$ is Hausdorff, every finite-dimensional vector subspace $M\subseteq L^p([0,1])$ is linearly homeomorphic to Euclidean space ${ℝ}^{\dim M}$ or ${ℂ}^{\dim M}$ (by F. Riesz's theorem) and so every non-zero linear functional $f$ on $M$ is continuous but none has a continuous linear extension to all of $L^p([0,1]).$ However, it is possible for a TVS $X$ to not be locally convex but nevertheless have enough continuous linear functionals that its continuous dual space $X^{*}$ separates points; for such a TVS, a continuous linear functional defined on a vector subspace Template:Em have a continuous linear extension to the whole space.

If the TVS $X$ is not locally convex then there might not exist any continuous seminorm $p:X\to ℝ$ Template:Em (not just on $M$) that dominates $f,$ in which case the Hahn–Banach theorem can not be applied as it was in the above proof of the continuous extension theorem. However, the proof's argument can be generalized to give a characterization of when a continuous linear functional has a continuous linear extension: If $X$ is any TVS (not necessarily locally convex), then a continuous linear functional $f$ defined on a vector subspace $M$ has a continuous linear extension $F$ to all of $X$ if and only if there exists some continuous seminorm $p$ on $X$ that dominates $f.$ Specifically, if given a continuous linear extension $F$ then $p:=|F|$ is a continuous seminorm on $X$ that dominates $f;$ and conversely, if given a continuous seminorm $p:X\to ℝ$ on $X$ that dominates $f$ then any dominated linear extension of $f$ to $X$ (the existence of which is guaranteed by the Hahn–Banach theorem) will be a continuous linear extension.

Geometric Hahn–Banach (the Hahn–Banach separation theorems)

The key element of the Hahn–Banach theorem is fundamentally a result about the separation of two convex sets: $\{-p(-x-n)-f(n):n\in M\},$ and $\{p(m+x)-f(m):m\in M\}.$ This sort of argument appears widely in convex geometry, optimization theory, and economics. Lemmas to this end derived from the original Hahn–Banach theorem are known as the Hahn–Banach separation theorems. They are generalizations of the hyperplane separation theorem, which states that two disjoint nonempty convex subsets of a finite-dimensional space ${ℝ}^n$ can be separated by some Template:Em, which is a fiber (level set) of the form $f^{-1}(s)=\{x:f(x)=s\}$ where $f\ne 0$ is a non-zero linear functional and $s$ is a scalar.

When the convex sets have additional properties, such as being open or compact for example, then the conclusion can be substantially strengthened:

Then following important corollary is known as the Geometric Hahn–Banach theorem or Mazur's theorem (also known as Ascoli–Mazur theorem). It follows from the first bullet above and the convexity of $M.$

Mazur's theorem clarifies that vector subspaces (even those that are not closed) can be characterized by linear functionals.

Supporting hyperplanes

Since points are trivially convex, geometric Hahn–Banach implies that functionals can detect the boundary of a set. In particular, let $X$ be a real topological vector space and $A\subseteq X$ be convex with $\mathrm{Int}A\ne \varnothing .$ If $a_0\in A\setminus \mathrm{Int}A$ then there is a functional that is vanishing at $a_0,$ but supported on the interior of $A.$

Call a normed space $X$ smooth if at each point $x$ in its unit ball there exists a unique closed hyperplane to the unit ball at $x.$ Köthe showed in 1983 that a normed space is smooth at a point $x$ if and only if the norm is Gateaux differentiable at that point.Template:Sfn

Balanced or disked neighborhoods

Let $U$ be a convex balanced neighborhood of the origin in a locally convex topological vector space $X$ and suppose $x\in X$ is not an element of $U.$ Then there exists a continuous linear functional $f$ on $X$ such thatTemplate:Sfn $\sup |f(U)|\le |f(x)|.$

Applications

The Hahn–Banach theorem is the first sign of an important philosophy in functional analysis: to understand a space, one should understand its continuous functionals.

For example, linear subspaces are characterized by functionals: if Template:Mvar is a normed vector space with linear subspace Template:Mvar (not necessarily closed) and if $z$ is an element of Template:Mvar not in the closure of Template:Mvar, then there exists a continuous linear map $f:X\to 𝐊$ with $f(m)=0$ for all $m\in M,$ $f(z)=1,$ and $\Vert f\Vert =\mathrm{dist}(z,M{)}^{-1}.$ (To see this, note that $\mathrm{dist}(\cdot ,M)$ is a sublinear function.) Moreover, if $z$ is an element of Template:Mvar, then there exists a continuous linear map $f:X\to 𝐊$ such that $f(z)=\Vert z\Vert$ and $\Vert f\Vert \le 1.$ This implies that the natural injection $J$ from a normed space Template:Mvar into its double dual $V^{**}$ is isometric.

That last result also suggests that the Hahn–Banach theorem can often be used to locate a "nicer" topology in which to work. For example, many results in functional analysis assume that a space is Hausdorff or locally convex. However, suppose Template:Mvar is a topological vector space, not necessarily Hausdorff or locally convex, but with a nonempty, proper, convex, open set Template:Mvar. Then geometric Hahn–Banach implies that there is a hyperplane separating Template:Mvar from any other point. In particular, there must exist a nonzero functional on Template:Mvar — that is, the continuous dual space $X^{*}$ is non-trivial.Template:SfnTemplate:Sfn Considering Template:Mvar with the weak topology induced by $X^{*},$ then Template:Mvar becomes locally convex; by the second bullet of geometric Hahn–Banach, the weak topology on this new space separates points. Thus Template:Mvar with this weak topology becomes Hausdorff. This sometimes allows some results from locally convex topological vector spaces to be applied to non-Hausdorff and non-locally convex spaces.

Partial differential equations

The Hahn–Banach theorem is often useful when one wishes to apply the method of a priori estimates. Suppose that we wish to solve the linear differential equation $Pu=f$ for $u,$ with $f$ given in some Banach space Template:Mvar. If we have control on the size of $u$ in terms of $\Vert f{\Vert }_X$ and we can think of $u$ as a bounded linear functional on some suitable space of test functions $g,$ then we can view $f$ as a linear functional by adjunction: $(f,g)=(u,P^{*}g).$ At first, this functional is only defined on the image of $P,$ but using the Hahn–Banach theorem, we can try to extend it to the entire codomain Template:Mvar. The resulting functional is often defined to be a weak solution to the equation.

Example from Fredholm theory

To illustrate an actual application of the Hahn–Banach theorem, we will now prove a result that follows almost entirely from the Hahn–Banach theorem.

The above result may be used to show that every closed vector subspace of ${ℝ}^{\mathbb{N}}$ is complemented because any such space is either finite dimensional or else TVS–isomorphic to ${ℝ}^{\mathbb{N}}.$

Generalizations

General template

There are now many other versions of the Hahn–Banach theorem. The general template for the various versions of the Hahn–Banach theorem presented in this article is as follows:

$p:X\to ℝ$ is a sublinear function (possibly a seminorm) on a vector space $X,$ $M$ is a vector subspace of $X$ (possibly closed), and $f$ is a linear functional on $M$ satisfying $|f|\le p$ on $M$ (and possibly some other conditions). One then concludes that there exists a linear extension $F$ of $f$ to $X$ such that $|F|\le p$ on $X$ (possibly with additional properties).

For seminorms

So for example, suppose that $f$ is a bounded linear functional defined on a vector subspace $M$ of a normed space $X,$ so its the operator norm $\Vert f\Vert$ is a non-negative real number. Then the linear functional's absolute value $p:=|f|$ is a seminorm on $M$ and the map $q:X\to ℝ$ defined by $q(x)=\Vert f\Vert \Vert x\Vert$ is a seminorm on $X$ that satisfies $p\le q{|}_M$ on $M.$ The Hahn–Banach theorem for seminorms guarantees the existence of a seminorm $P:X\to ℝ$ that is equal to $|f|$ on $M$ (since $P{|}_M=p=|f|$) and is bounded above by $P(x)\le \Vert f\Vert \Vert x\Vert$ everywhere on $X$ (since $P\le q$).

Maximal dominated linear extension

If $S=\{s\}$ is a singleton set (where $s\in X$ is some vector) and if $F:X\to ℝ$ is such a maximal dominated linear extension of $f:M\to ℝ,$ then $F(s)=\underset{m\in M}{\inf }[f(s)+p(s-m)].$Template:Sfn

Invariant Hahn–Banach

A set $\mathrm{\Gamma }$ of maps $X\to X$ is Template:Em (with respect to function composition $\circ$) if $F\circ G=G\circ F$ for all $F,G\in \mathrm{\Gamma }.$ Say that a function $f$ defined on a subset $M$ of $X$ is Template:Em if $L(M)\subseteq M$ and $f\circ L=f$ on $M$ for every $L\in \mathrm{\Gamma }.$

This theorem may be summarized:

Every $\mathrm{\Gamma }$-invariant continuous linear functional defined on a vector subspace of a normed space $X$ has a $\mathrm{\Gamma }$-invariant Hahn–Banach extension to all of $X.$Template:Sfn

For nonlinear functions

The following theorem of Mazur–Orlicz (1953) is equivalent to the Hahn–Banach theorem.

The following theorem characterizes when Template:Em scalar function on $X$ (not necessarily linear) has a continuous linear extension to all of $X.$

Converse

Let Template:Mvar be a topological vector space. A vector subspace Template:Mvar of Template:Mvar has the extension property if any continuous linear functional on Template:Mvar can be extended to a continuous linear functional on Template:Mvar, and we say that Template:Mvar has the Hahn–Banach extension property (HBEP) if every vector subspace of Template:Mvar has the extension property.Template:Sfn

The Hahn–Banach theorem guarantees that every Hausdorff locally convex space has the HBEP. For complete metrizable topological vector spaces there is a converse, due to Kalton: every complete metrizable TVS with the Hahn–Banach extension property is locally convex.Template:Sfn On the other hand, a vector space Template:Mvar of uncountable dimension, endowed with the finest vector topology, then this is a topological vector spaces with the Hahn–Banach extension property that is neither locally convex nor metrizable.Template:Sfn

A vector subspace Template:Mvar of a TVS Template:Mvar has the separation property if for every element of Template:Mvar such that $x\in ̸M,$ there exists a continuous linear functional $f$ on Template:Mvar such that $f(x)\ne 0$ and $f(m)=0$ for all $m\in M.$ Clearly, the continuous dual space of a TVS Template:Mvar separates points on Template:Mvar if and only if $\{0\},$ has the separation property. In 1992, Kakol proved that any infinite dimensional vector space Template:Mvar, there exist TVS-topologies on Template:Mvar that do not have the HBEP despite having enough continuous linear functionals for the continuous dual space to separate points on Template:Mvar. However, if Template:Mvar is a TVS then Template:Em vector subspace of Template:Mvar has the extension property if and only if Template:Em vector subspace of Template:Mvar has the separation property.Template:Sfn

Relation to axiom of choice and other theorems

Template:AnchorThe proof of the Hahn–Banach theorem for real vector spaces (HB) commonly uses Zorn's lemma, which in the axiomatic framework of Zermelo–Fraenkel set theory (ZF) is equivalent to the axiom of choice (AC). It was discovered by Łoś and Ryll-NardzewskiTemplate:Sfn and independently by LuxemburgTemplate:Sfn that HB can be proved using the ultrafilter lemma (UL), which is equivalent (under ZF) to the Boolean prime ideal theorem (BPI). BPI is strictly weaker than the axiom of choice and it was later shown that HB is strictly weaker than BPI.Template:Sfn

The ultrafilter lemma is equivalent (under ZF) to the Banach–Alaoglu theorem,Template:Sfn which is another foundational theorem in functional analysis. Although the Banach–Alaoglu theorem implies HB, it is not equivalent to it (said differently, the Banach–Alaoglu theorem is strictly stronger than HB). However, HB is equivalent to a certain weakened version of the Banach–Alaoglu theorem for normed spaces. The Hahn–Banach theorem is also equivalent to the following statement:

(∗): On every Boolean algebra Template:Mvar there exists a "probability charge", that is: a non-constant finitely additive map from $B$ into $[0,1].$

(BPI is equivalent to the statement that there are always non-constant probability charges which take only the values 0 and 1.)

In ZF, the Hahn–Banach theorem suffices to derive the existence of a non-Lebesgue measurable set. Moreover, the Hahn–Banach theorem implies the Banach–Tarski paradox.

For separable Banach spaces, D. K. Brown and S. G. Simpson proved that the Hahn–Banach theorem follows from WKL₀, a weak subsystem of second-order arithmetic that takes a form of Kőnig's lemma restricted to binary trees as an axiom. In fact, they prove that under a weak set of assumptions, the two are equivalent, an example of reverse mathematics.

See also

• Template:Annotated link
• Template:Annotated link
• Template:Annotated link
• Template:Annotated link
• Template:Annotated link

Notes

Proofs

Bibliography

• Template:Adasch Topological Vector Spaces
• Template:Banach Théorie des Opérations Linéaires
• Template:Berberian Lectures in Functional Analysis and Operator Theory
• Template:Bourbaki Topological Vector Spaces
• Template:Conway A Course in Functional Analysis
• Template:Edwards Functional Analysis Theory and Applications
• Template:Grothendieck Topological Vector Spaces
• Template:Jarchow Locally Convex Spaces
• Template:Köthe Topological Vector Spaces I
• Template:Cite journal
• Template:Cite journal
• Template:Cite book
• Template:Cite journal
• Template:Narici Beckenstein Topological Vector Spaces
• Template:Cite journal
• Template:Cite book
• Reed, Michael and Simon, Barry, Methods of Modern Mathematical Physics, Vol. 1, Functional Analysis, Section III.3. Academic Press, San Diego, 1980. Template:ISBN.
• Template:Cite book
• Template:Robertson Topological Vector Spaces
• Template:Rudin Walter Functional Analysis
• Template:Schaefer Wolff Topological Vector Spaces
• Template:Cite journal
• Template:Schechter Handbook of Analysis and Its Foundations
• Template:Swartz An Introduction to Functional Analysis
• Tao, Terence, The Hahn–Banach theorem, Menger's theorem, and Helly's theorem
• Template:Trèves François Topological vector spaces, distributions and kernels
• Template:Wilansky Modern Methods in Topological Vector Spaces
• Wittstock, Gerd, Ein operatorwertiger Hahn-Banach Satz, J. of Functional Analysis 40 (1981), 127–150
• Zeidler, Eberhard, Applied Functional Analysis: main principles and their applications, Springer, 1995.
• Template:Springer