logo

Dense Subsets and Closures 📂Banach Space

Dense Subsets and Closures

Dense Subsets

Definition1

Let WVW \subset V be a subset of the normed space VV. For any vV\mathbf{v} \in V and ϵ>0\epsilon \gt 0, if there always exists wW\mathbf{w} \in W satisfying the following, then WW is called a dense subset in VV.

vwϵ \left\| \mathbf{v} - \mathbf{w} \right\| \le \epsilon

Explanation

If WW is a dense subspace of VV, it implies that any element within VV can be well approximated by some element in WW.

Let vV\mathbf{v} \in V and assume ϵ=1k(kN)\epsilon = \dfrac{1}{k} (k \in \N). If WW is a dense subset of VV, by definition, there exists wkW\mathbf{w}_{k} \in W satisfying the following.

vwk1k \left\| \mathbf{v} - \mathbf{w}_{k} \right\| \le \dfrac{1}{k}

Therefore, the sequence {wk}\left\{ \mathbf{w}_{k} \right\} converges to v\mathbf{v} when kk \to \infty.

limkwk=v \lim \limits_{k \to \infty} \mathbf{w}_{k} = \mathbf{v}

Thus, saying WW is a dense subset in VV means there exists a sequence {wk}\left\{ \mathbf{w}_{k} \right\} of WW converging to vV\mathbf{v} \in V.

Closure

Definition

Let WW be a subset of the normed space VV. For a given ϵ>0\epsilon \gt 0, the set of all vV\mathbf{v} \in V for which there exists wW\mathbf{w} \in W satisfying vwϵ\left\| \mathbf{v} - \mathbf{w} \right\| \le \epsilon is defined as the closure of WW and is denoted by W\overline{W}.

W:={vVfor each ϵ>0,wWs.t. vwϵ} \overline{W} := \left\{ \mathbf{v} \in V | \text{for each } \epsilon \gt 0, \exist \mathbf{w} \in W \text{s.t. } \left\| \mathbf{v} - \mathbf{w} \right\| \le \epsilon\right\}

Explanation

It can be seen as the opposite definition of a dense subset. Since all wWV\mathbf{w}\in W\subset V satisfy ww0\left\| \mathbf{w} - \mathbf{w} \right\| \le 0, naturally, WWW \subset \overline{W} holds.

Theorem

Let WW be a subset of the normed space VV. Then, WW being dense in VV is equivalent to W=V\overline{W} = V.

Proof

It is trivial by the definitions of dense subsets and closure.


  1. Ole Christensen, Functions, Spaces, and Expansions: Mathematical Tools in Physics and Engineering (2010), p35-36 ↩︎