Function spaces and compactness

Vaughn Climenhaga
February 13, 2013

In the last two posts on spectral methods in dynamics, we’ve used (both
explicitly and implicitly) a number of results and a good deal of intuition on
function spaces. It seems worth discussing these a little more at length, as a
supplement to the weekly seminar posting.

1 Function spaces and extra structure

It is useful to treat real-valued functions (or complex-valued functions, or
vector space-valued functions) as elements of a vector space, so that the
tools from linear algebra can be applied. Given a set X one may consider
the vector space RX of all real-valued functions with domain X. If X is finite,
say with n elements, then this is just the familiar vector space Rn . The more
interesting examples are when X is infinite, and so RX is infinite-dimensional.
We will focus on the case X = [0, 1], which is reasonably representative.
Generally speaking, the functions [0, 1] → R that arise from some appli-
cation are not entirely arbitrary, but have some degree of regularity – maybe
they are continuous, or piecewise continuous, or measurable, or integrable,
etc. It turns out that the vector space R[0,1] is “too large” for many ap-
plications, and that it is more suitable to consider a smaller space, whose
elements are functions with some extra properties. We will consider some of
the ways to do this, paying particular attention to how those choices let us
recover certain properties of Rn that involve extra structure beyond that of
the vector space itself:
• Topology: We know what it means for a sequence ~xk ∈ Rn to converge
to some ~x ∈ Rn , and we want a similar notion of convergence in a
vector space V ⊂ R[0,1] .

1
 • Metric and norm: We want the notion of convergence to come from
a metric (distance function) that is compatible with the vector space
structure of V – that is, a norm, with respect to which the vector space
V becomes a Banach space.

• Compactness: A subset of Rn is compact if every sequence in that sub-

set has a convergent subsequence, and this property is important in
many applications and proofs. By the Heine–Borel theorem compact-
ness in Rn is equivalent to being closed and bounded. How can we
determine when a set of functions in V is compact?

2 Continuous functions and Arzelà–Ascoli

The extra structure we seek to place on V ⊂ R[0,1] should leverage some of
the extra structure that [0, 1] has, beyond simply being an uncountable set.
In particular, we may use either the topology of [0, 1] or Lebesgue measure
on [0, 1] to define properties of functions f : [0, 1] → R. First we discuss the
topological option – later we see what happens when we use the measure-
theoretic structure to define the Lp spaces (and others).
The natural space to use is C(X), the space of continuous real-valued
functions on X = [0, 1], with the norm kf kC 0 = supx∈[0,1] |f (x)|. The space
of continuous functions is complete with respect to this norm, and so we
have a Banach space. What about compactness? How do we tell if a set
A ⊂ C(X) is compact? Of course A should be closed, but what else do
we need? Boundedness is no longer enough: the unit ball in C(X) is not
compact, as can be seen by considering the sequence of functions shown in
Figure 1.
The solution here is given by the Arzelà–Ascoli theorem: a set A ⊂ C(X)
is pre-compact (has compact closure) if and only if the following conditions
are satisfied.

• A is uniformly bounded: supf ∈A supx∈X |f (x)| < ∞.

• A is equicontinuous: for every ε > 0 there exists δ > 0 such that

|f (x) − f (y)| < ε for every f ∈ A and |x − y| < δ.

Remark 1 The proof that these conditions guarantee compactness uses the
following strategy, which it is a useful exercise to complete:

2
 Figure 1: Uniformly bounded but no convergent subsequence.

1. Given any sequence fn ∈ A, use uniform boundedness and a diagonali-

sation argument to find a subsequence that converges at every rational
number. (Or on some other countable dense set.)

2. Use equicontinuity to guarantee that {fnk (x)}k≥1 is Cauchy for every

x ∈ [0, 1], and hence converges.

In particular, one can consider the subspace C α (X) ⊂ C(X) of Hölder

continuous functions with exponent α ∈ (0, 1) – this is a Banach space with
norm
|f (x) − f (y)|
kf kC α = kf kC 0 + |f |α , |f |α = sup .
x6=y |x − y|α
When α = 1 this is the space of Lipschitz functions. If A ⊂ C α (X) is
uniformly bounded in the C α norm, then it is uniformly bounded in the C 0
norm and equicontinuous, and hence it is pre-compact in the C 0 norm.
It is important to note here the structure of the last statement – we
have two norms, k · kC α and k · kC 0 , such that uniform boundedness in
one norm implies pre-compactness in the other. This is the closest
that we can come to an infinite dimensional analogue of Heine–Borel: as a
consequence of Riesz’s lemma, every infinite-dimensional Banach space has
a uniformly bounded sequence with no convergent subsequence.
In our study of spectral methods in dynamics, an important step is always
to find two norms with this relationship: uniform boundedness in one implies
pre-compactness in the other. We remark that the Arzelà–Ascoli theorem
actually gives just a little bit more than this: given a sequence fn ∈ C(X)

3
that is uniformly bounded in the C α norm, pre-compactness only guarantees
C0
the existence of a limit point fnk −→ f ∈ C 0 , but in fact the limit point f is
in C α as well, because any modulus of continuity for the sequence fn is also
a modulus of continuity for any limit point.
Another important family of function spaces, which leverages not only
the topological but also the differentiable structure of the unit interval, are
the spaces C r , defined inductively as

C r+1 = {f : [0, 1] → R | f is differentiable and f 0 ∈ C r }.

Here r need not be an integer (the base case for the induction is 0 ≤ r < 1),
so for example, for 0 < α < 1, C 1+α is the space of differentiable functions
whose derivatives are Hölder continuous with exponent α. The space C r
becomes a Banach space when endowed with the norm inductively given by

kf kC r+1 = kf kC 0 + kf 0 kC r .

For example, on C 1 the appropriate norm is

kf kC 1 = kf kC 0 + kf 0 kC 0 . (1)

The relationship discussed above between uniform boundedness in one norm

and pre-compactness in another can be stated quite generally for this fam-
ily of norms: uniform boundedness in the C r norm implies pre-
compactness in the C s norm for any 0 ≤ s < r. This relationship is
often expressed by saying that “C r is compactly embedded in C s for r > s”.

3 Lp spaces
In terms of the measure-theoretic structure of the unit interval, the most
important function spaces are the Lp spaces

Lp = Lp ([0, 1], dx)

(

= f : [0, 1] → R  f is measurable and
Z  p1 )
kf kp := |f (x)|p dx <∞ ,
[0,1]

4
where 1 ≤ p < ∞, and

L∞ = {f : [0, 1] → R | f is measurable and kf k∞ < ∞},

where kf k∞ = sup{L ≥ 0 | {x ∈ [0, 1] | |f (x)| > L} has positive Lebesgue measure}

is the essential supremum of f .
In fact, this definition cheats a little bit, because elements of an Lp space
are actually equivalence classes of functions, where two functions are equiv-
alent if they agree on a set of full Lebesgue measure. This throws a small
technical monkey wrench into many arguments involving Lp spaces, since
strictly speaking an expression like f (x) for f ∈ Lp has no meaning unless
it is inside an integral sign. One way to avoid these technicalities is to em-
phasise the role of elements of Lp not necessary as functions, but rather as
linear functionals.
Recall that if B is a Banach space, then B ∗ is the dual space of continuous
linear functionals B → R. The Lp spaces have the property that
1 1
(Lp )∗ = Lq for 1 < p, q < ∞ such that + = 1,
p q
where f ∈ Lp defines a linear functional on Lq by
Z
g 7→ f · g dx for g ∈ Lq . (2)

Thus instead of thinking of a function f ∈ Lp , we may think of the associated

functional in (2), which is obtained by integrating the function f against test
functions from a suitable space. In this case the space of test functions
is taken to be Lq , but there are many other examples we could consider
– eventually this leads to the idea of considering distributions in place of
functions, but we will not go this far here.

Remark 2 Before moving on, we note that (L1 )∗ = L∞ , but (L∞ )∗ is a

larger space than L1 .

4 Weak derivatives
An important use of this alternate viewpoint – functions as continuous linear
functionals – is to define the weak derivative of a function. If f : [0, 1] → R is

5
differentiable, then for any differentiable g : [0, 1] → R with g(0) = g(1) = 0,
integration by parts gives
Z Z
f · g dx = − f · g 0 dx.
0
(3)

Equation (3) characterises the derivative f 0 , which motivates the following

definition: h ∈ L1 is the weak derivative of f ∈ L1 if
Z Z
h · ϕ dx = − f · ϕ0 dx for all ϕ ∈ G, (4)

where the space of test functions is G = {ϕ ∈ C 1 ([0, 1], R) | ϕ(0) = ϕ(1) =

0}. Write h = Df in this case.

Example 1 The absolute value function f (x) = |x| has as its derivative the
step function Df (x) = −1(x < 0), 1(x > 0). Note that the value of Df (0) is
not uniquely defined because Df is considered as an element of L1 .
Writing g(x) = Df (x) for the step function just described, we see that g
does not have a weak derivative in L1 . Indeed, this is true for any function
with a jump discontinuity.

Using mollifiers one can show that any L1 function f can be L1 approx-
imated by (infinitely) differentiable functions f such that f0 approximates
Df in L1 . This can be used to show that the usual product rule for deriva-
tives holds for weak derivatives as well: D(f g) = (Df ) · g + f · (Dg), as
long as f and g both have weak derivatives. The space of L1 functions with
a weak derivative in L1 is denoted W 1,1 and is an important example of a
Sobolev space. Here the norm is

kf kW 1,1 = kf kL1 + kDf kL1 ,

which can be viewed as an analogue of the definition of the C 1 norm in (1).

Moreover, just as the C 1 unit ball is C 0 compact, so also the W 1,1 unit ball
is L1 compact, as we will see.

5 Kolmogorov–Riesz compactness theorem

In understanding compactness for subsets of function spaces, it is useful to
recall that the Heine–Borel theorem can be generalised to arbitrary complete

6
metric spaces as follows: a set is compact if and only if it is closed and totally
bounded. In particular, for Banach spaces, pre-compactness is equivalent to
being totally bounded.
The Arzelà–Ascoli theorem gives a necessary and sufficient condition for a
set in C 0 to be totally bounded (and hence pre-compact). A similar result in
the Lp spaces is the Kolmogorov–Riesz compactness theorem – an expository
account of this theorem and its relationship to the Arzelà–Ascoli theorem is
given in a recent paper by H. Hanche–Olsen and H. Holden, The Kolmogorov–
Riesz compactness theorem (Expo. Math. 28 (2010), 385–394).
In our setting (where we consider Lp spaces with respect to a finite mea-
sure), the Kolmogorov–Riesz theorem can be stated as follows: a set F ⊂ Lp
is totally bounded (in the Lp norm) if and only if

1. F is bounded, and

2. for every ε > 0 there is δ > 0 such that kf ◦ Tγ − f kp < ε for every
f ∈ F and |γ| < δ, where Tγ : x 7→ x + γ.

In other words, to go from bounded to totally bounded one needs the added
condition that small changes to the argument result in (uniformly) small
changes in the function, with respect to the Lp norm.
Roughly speaking the idea is that if a set can be “approximately embed-
ded” into a totally bounded set, then it must itself be totally bounded – this
is Lemma 1 in the paper referred to above. Then the condition on f ◦ Tρ − f
for f ∈ F allows the set F to be “approximately embedded” into a bounded
set in Rn by averaging f over small neighbourhoods in its domain. This is of
course a very rough description and one should read the paper for the com-
plete proof and precise formulation of what it means to be “approximately
embedded”.

6 Bounded variation and Helly’s theorem

One can use the Kolmogorov–Riesz theorem to show that W 1,1 is compactly
embedded in L1 . (This is a special case of the Rellich–Kondrachov theorem.)
However, since functions with jump discontinuities are not in W 1,1 , we want
to use a bigger function space in order to study spectral properties of the
transfer operator.

7
 The definition of weak derivative can be generalised if one is willing to
allow Df to live somewhere besides L1 . Recall that we want Df to satisfy
Z Z
(Df ) · ϕ dx = − f · ϕ0 dx

for every test function ϕ ∈ G, the space of C 1 functions on the interval

that vanish at the endpoints. The left-hand side defines a linear functional
G → R, and given any f ∈ L1 we may define Df as such a linear functional
by setting Z
(Df )(ϕ) = − f · ϕ0 dx.

/ W 1,1 , this functional is not given by integration against an L1 function,

If f ∈
but now the definition makes sense for any f ∈ L1 . Moreover, the space of
linear functionals on G carries a natural norm: the norm of ` : G → R is
k`kG ∗ = sup{|`(ϕ)| | ϕ ∈ G, kϕkC 0 ≤ 1}.
A functional ` is continuous if and only if k`k < ∞. Recalling our discussion
of bounded variation functions in an earlier post, we see that kDf kG ∗ =
|f |BV , and so
BV = {f ∈ L1 | kDf kG ∗ < ∞}.
The BV norm can be written as kf kBV = kf kL1 + kDf kG ∗ . Note that BV
is exactly the set of functions f ∈ L1 for which Df is a continuous linear
functional on G.
Helly’s selection theorem states that BV is compactly embedded in L1 .
(This is not to be confused with Helly’s theorem in geometry.) This is a con-
sequence of the Kolmogorov–Riesz compactness theorem, because a relatively
straightforward computation shows that
kf ◦ Tγ − f kL1 ≤ |f |BV |γ|.
(See Lemma 11 and Theorem 12 in the paper of Hanche–Olsen and Holden
referenced above.) We remark that one can also give a direct proof following
the hint given in Footnote 8 of Keller and Liverani’s A spectral gap for a
one-dimensional lattice of coupled piecewise expanding interval maps: given
f ∈ BV , let fn be the step function that is constant on each dyadic interval
[k, k + 1]/2n , with value equal to the average of f on that interval. Then the
functions fn approach f in L1 , and the problem reduces to finding a suitable
subsequence of step functions.