logo

L1 Space 📂Lebesgue Spaces

L1 Space

Definition1

A function space $L^{1}$ is defined as follows.

$$ L^{1} (E) := \left\{ f : E \to \mathbb{R} \Big \vert \int_{E} | f | dm \lt \infty \right\} $$

Properties

  1. $L^{1}$ is a vector space.
  2. $L^{1}$ is a normed space. The norm is defined as follows: $$ \left\| f \right\|_{1} := \int \left| f(x) \right| dx $$
  3. $L^{1}$ is a complete space.

Explanation

Space $L^{1}$ is a special case when it is the $p=1$ of the $L^{p}$ space, and has been defined as a collection of integrable functions when talking about Lebesgue integrability.

Generalized proofs for the $L^{p}$ space can be found here.

Proof

2.

Definition of Norm

Let’s say $V$ in $\mathbb{F}$ is a vector space. If a function $\left\| \cdot \right\| : V \to \mathbb{F}$ satisfies the following three conditions for $\mathbf{u}, \mathbf{v} \in V$ and $k \in \mathbb{F}$, then $\left\| \cdot \right\|$ is defined as the norm on $V$.

  • Positive definiteness: $\left\| \mathbf{u} \right\| \ge 0$ and $\mathbf{u} = \mathbb{0} \iff \left\| \mathbf{u} \right\| = 0$
  • Homogeneity: $\left\|k \mathbf{u} \right\| = | k | \left\| \mathbf{u} \right\| $
  • Triangle inequality: $\left\| \mathbf{u} + \mathbf{v}\right\| \le \left\|\mathbf{v} \right\| + \left\| \mathbf{u} \right\|$

Let’s define the norm of $L ^{1}$ as $\displaystyle \left\| f \right\|_{1} := \int_{E} |f| dm$.

  • Part 1. Positive Definiteness

    Since $| f | \ge 0$, if $f = 0$ almost everywhere then $\left\| f \right\|_{1} = 0$. Conversely, if $\left\| f \right\|_{1} = 0$ then almost everywhere must be $f = 0$.

  • Part 2. Homogeneity

    $$\left\| c f \right\| _{1} = \int_{E} | c f | dm = |c| \int_{E} | f | dm = |c| \left\| f \right\| _{1}$$

  • Part 3. Triangle Inequality

    $$ \left\| f + g \right\|_{1} = \int_{E} | f + g | dm \le \int_{E} | f | dm + \int_{E} | g | dm = \left\| f\right\|_{1} + \left\| g\right\|_{1} $$

3.

Completeness

Let’s assume that the norm $\left\| \cdot \right\|_{X}$ is defined in the vector space $X$. If for every $\varepsilon > 0$ $$n, m \ge N \implies \left\| f_{n} - f_{m} \right\|_{X} \lt \varepsilon$$ If there exists a $N \in \mathbb{N}$ satisfying this, then the sequence $f_{n} \in X$ is called a Cauchy sequence. If every Cauchy sequence converges to an element of $X$, then $X$ is called complete.

If $f_{n} \in L^{1}$ is a Cauchy sequence,

$$ \left\| f_{n} - f_{N_{1}} \right\|_{1} \lt {{1} \over {2}} $$

$N_{1}$ can be found that satisfies this, and similarly,

$$ \left\| f_{n} - f_{N_{2}} \right\|_{1} \lt {{1} \over {2^2}} $$

$N_{2} > N_{1}$ can be found that satisfies this. In this way, $$ \left\| f_{n} - f_{N_{n}} \right\|_{1} \lt {{1} \over {2^n}} $$ $N_{n} > N_{n-1}$ can be found that satisfies this. By the triangle inequality, $$ \left\| f_{N_{n}} - f_{N_{n-1}} \right\|_{1} \lt \left\| f_{N_{n}} - f_{n} \right\|_{1} + \left\| f_{n} - f_{N_{n-1}} \right\|_{1} \lt {{1} \over {2^n}} + {{1} \over {2^{n-1}}} \lt {{3} \over {2^{n}}} $$

Levi’s Theorem

If $\displaystyle \sum_{k=1}^{\infty} \int |f_{k}| dm \lt \infty$ then $\displaystyle \sum_{k=1}^{\infty} f_{k} (x)$ converges almost everywhere and the following holds:

$$ \int \sum_{k=1}^{\infty} f_{k} dm = \sum_{k=1}^{\infty} \int f_{k} dm $$

By Levi’s Theorem, $\displaystyle \sum_{n=1}^{\infty} | f_{N_{n}} - f_{N_{n-1}} |_{1}$ converges. Therefore, the following converges almost everywhere.

$$ f_{N_{1}}(x) + \sum_{n=2}^{ k } \left[ f_{N_{n}} (x) - f_{N_{n-1}} (x) \right] = f_{N_{k}} $$

If the right side converges to $f(x)$, then the right side’s $f_{N_{k}} (x)$ also converges to $f(x)$.

Fatou’s Lemma

For a sequence of non-negative measurable functions $\left\{ f_{n} \right\}$,

$$ \displaystyle \int_{E} \left( \liminf_{n \to \infty} f_{n} \right) dm \le \liminf_{n \to \infty} \int_{E} f_{n} dm $$

By Fatou’s Lemma,

$$ \begin{align*} \left\| f - f_{n} \right\|_{1} =& \int |f - f_{n}| dm \\ \le & \liminf_{k \to \infty} \int | f_{N_{k}} - f_{n}| dm \\ =& \liminf_{k \to \infty} \left\| f_{N_{k}} - f_{n} \right\| \\ \lt& \varepsilon \end{align*} $$

Since $f_{n}$ is a Cauchy sequence, the above inequality holds for any $\varepsilon > 0$, hence $\left\| f_{n} - f \right\|_{1} \to 0$ is obtained. In short, since $f_{n}$ is Cauchy and a subsequence converges to $f$, $f_{n}$ converges to $f$. Here, since $f - f_{n} \in L^{1}$ and $L^{1}$ is a vector space,

$$ ( f - f_{n} ) + f_{n} = f \in L^{1} $$

All Cauchy sequences in $L^{1}$ converge to an element of $L^{1}$, hence $L^{1}$ is a complete space.

See Also


  1. Capinski. (1999). Measure, Integral and Probability: p127. ↩︎