Difference between revisions of "Baire's Theorem and Applications"

From Department of Mathematics at UTSA
Jump to navigation Jump to search
(Created page with "<h1 id="toc0"><span>The Baire Category Theorem</span></h1> <blockquote style="background: white; border: 1px solid black; padding: 1em;"> <td><strong>Lemma 1:</strong> Let <sp...")
 
 
(4 intermediate revisions by the same user not shown)
Line 1: Line 1:
<h1 id="toc0"><span>The Baire Category Theorem</span></h1>
+
==Dense and Nowhere Dense Sets==
 +
===Dense Sets in a Topological Space===
 +
<blockquote style="background: white; border: 1px solid black; padding: 1em;">
 +
<td><strong>Definition:</strong> Let <span class="math-inline"><math>(X, \tau)</math></span> be a topological space. The set <span class="math-inline"><math>A \subseteq X</math></span> is said to be <strong>Dense</strong> in <span class="math-inline"><math>X</math></span> if the intersection of every nonempty open set with <span class="math-inline"><math>A</math></span> is nonempty, that is, <span class="math-inline"><math>A \cap U \neq \emptyset</math></span> for all <span class="math-inline"><math>U \in \tau \setminus \{ \emptyset \}</math></span>.</td>
 +
</blockquote>
 +
<p>Given any topological space <span class="math-inline"><math>(X, \tau)</math></span> it is important to note that <span class="math-inline"><math>X</math></span> is dense in <span class="math-inline"><math>X</math></span> because every <span class="math-inline"><math>U \in \tau</math></span> is such that <span class="math-inline"><math>U \subseteq X</math></span>, and so <span class="math-inline"><math>X \cap U = U \neq \emptyset</math></span> for all <span class="math-inline"><math>U \in \tau \setminus \{ \emptyset \}</math></span>.</p>
 +
<p>For another example, consider the topological space <span class="math-inline"><math>(\mathbb{R}, \tau)</math></span> where <span class="math-inline"><math>\tau</math></span> is the usual topology of open intervals. Then the set of rational numbers <span class="math-inline"><math>\mathbb{Q} \subset \mathbb{R}</math></span> is dense in <span class="math-inline"><math>\mathbb{R}</math></span>. If not, then there exists an <span class="math-inline"><math>U \in \tau \setminus \{ \emptyset \}</math></span> such that <span class="math-inline"><math>\mathbb{Q} \cap U = \emptyset</math></span>.</p>
 +
<p>Since <span class="math-inline"><math>U \in \tau</math></span> we have that <span class="math-inline"><math>(a, b) \subseteq U</math></span> for some open interval <span class="math-inline"><math>(a, b)</math></span> with <span class="math-inline"><math>a, b \in \mathbb{R}</math></span> and <span class="math-inline"><math>a < b</math></span>. Suppose that <span class="math-inline"><math>\mathbb{Q} \setminus U = \emptyset</math></span>. Then we must also have that:</p>
 +
<div style="text-align: center;"><math> \begin{align} \quad \mathbb{Q} \cap U = \mathbb{Q} \cap (a, b) = \emptyset \end{align}</math></div>
 +
<p>The intersection above implies that there exists no rational numbers in the interval <span class="math-inline"><math>(a, b)</math></span>, i.e., there exists no <span class="math-inline"><math>q \in \mathbb{Q}</math></span> such that <span class="math-inline"><math>a < q < b</math></span>. But this is a contradiction since for all <span class="math-inline"><math>a, b \in \mathbb{R}</math></span> with <span class="math-inline"><math>a < b</math></span> there ALWAYS exists a rational number <span class="math-inline"><math>q \in \mathbb{Q}</math></span> such that <span class="math-inline"><math>a < q < b</math></span>, i.e., <span class="math-inline"><math>q \in (a, b)</math></span>. So <span class="math-inline"><math>\mathbb{Q} \cap (a, b) \neq \emptyset</math></span> for all <span class="math-inline"><math>U \in \tau \setminus \{ \emptyset \}</math></span>. Thus, <span class="math-inline"><math>\mathbb{Q}</math></span> is dense in <span class="math-inline"><math>\mathbb{R}</math></span>.</p>
 +
<p>We will now look at a very important theorem which will give us a way to determine whether a set <span class="math-inline"><math>A \subseteq X</math></span> is dense in <span class="math-inline"><math>X</math></span> or not.</p>
 +
<blockquote style="background: white; border: 1px solid black; padding: 1em;">
 +
<td><strong>Theorem 1:</strong> Let <span class="math-inline"><math>(X, \tau)</math></span> be a topological space and let <span class="math-inline"><math>A \subseteq X</math></span>. Then <span class="math-inline"><math>A</math></span> is dense in <span class="math-inline"><math>X</math></span> if and only if <span class="math-inline"><math>\bar{A} = X</math></span>.</td>
 +
</blockquote>
 +
<ul>
 +
<li><strong>Proof:</strong> <span class="math-inline"><math>\Rightarrow</math></span> Suppose that <span class="math-inline"><math>A</math></span> is dense in <span class="math-inline"><math>X</math></span>. Then for all <span class="math-inline"><math>U \in \tau \setminus \{ \emptyset \}</math></span> we have that <span class="math-inline"><math>A \cap U = \emptyset</math></span>. Clearly <span class="math-inline"><math>\bar{A} \subseteq X</math></span> so we only need to show that <span class="math-inline"><math>X \subseteq \bar{A}</math></span>.</li>
 +
</ul>
 +
===Nowhere Dense Sets in a Topological Space===
 +
<blockquote style="background: white; border: 1px solid black; padding: 1em;">
 +
<td><strong>Definition:</strong> Let <span class="math-inline"><math>(X, \tau)</math></span> be a topological space. A set <span class="math-inline"><math>A \subseteq X</math></span> is said to be <strong>Nowhere Dense</strong> in <span class="math-inline"><math>X</math></span> if the interior of the closure of <span class="math-inline"><math>A</math></span> is empty, that is, <span class="math-inline"><math>\mathrm{int} (\bar{A}) = \emptyset</math></span>.</td>
 +
</blockquote>
 +
<p>For example, consider the topological space <span class="math-inline"><math>(\mathbb{R}, \tau)</math></span> where <span class="math-inline"><math>\tau</math></span> is the usually topology of open intervals on <span class="math-inline"><math>\mathbb{R}</math></span>, and consider the set of integers <span class="math-inline"><math>\mathbb{Z}</math></span>. The closure of <span class="math-inline"><math>\mathbb{Z}</math></span>, <span class="math-inline"><math>\bar{\mathbb{Z}}</math></span> is the smallest closed set containing <span class="math-inline"><math>\mathbb{Z}</math></span>. The smallest closed set containing <span class="math-inline"><math>\mathbb{Z}</math></span> is <span class="math-inline"><math>\mathbb{Z}</math></span> since <span class="math-inline"><math>\mathbb{Z}^c</math></span> is open as <span class="math-inline"><math>\mathbb{Z}^c</math></span> is an arbitrary union of open sets:</p>
 +
<div style="text-align: center;"><math> \begin{align} \quad \mathbb{Z}^c = ... (-2, -1) \cup (-1, 0) \cup (0, 1) \cup (1, 2) \cup ... \end{align}</math></div>
 +
<p>So what is the interior of <span class="math-inline"><math>\bar{\mathbb{Z}} = \mathbb{Z}</math></span>? It is the largest open set contained in <span class="math-inline"><math>\bar{\mathbb{Z}} = \mathbb{Z}</math></span>. All open sets of <span class="math-inline"><math>\mathbb{R}</math></span> with respect to this topology <span class="math-inline"><math>\tau</math></span> are either the empty set, an open interval, a union of open intervals, or the whole set (the union of all open intervals). But no open intervals are contained in <span class="math-inline"><math>\mathbb{Z}</math></span> and so:</p>
 +
<div style="text-align: center;"><math> \begin{align} \quad \mathrm{int} (\bar{\mathbb{Z}}) = \emptyset \end{align}</math></div>
 +
<p>Therefore <span class="math-inline"><math>\mathbb{Z}</math></span> is a nowhere dense set in <span class="math-inline"><math>\mathbb{R}</math></span> with respect to the usual topology <span class="math-inline"><math>\tau</math></span> on <span class="math-inline"><math>\mathbb{R}</math></span>.</p>
 +
 
 +
 
 +
==Sets of the First and Second Categories in a Topological Space==
 +
<p>Recall that if <span class="math-inline"><math>(X, \tau)</math></span> is a topological space then a set <span class="math-inline"><math>A \subseteq X</math></span> is said to be dense in <span class="math-inline"><math>X</math></span> if the intersection of <span class="math-inline"><math>A</math></span> with all open sets (except for the empty set) is nonempty, that is, for all <span class="math-inline"><math>U \in \tau \setminus \{ \emptyset \}</math></span> we have that:</p>
 +
<div style="text-align: center;"><math>\begin{align} \quad A \cap U \neq \emptyset \end{align}</math></div>
 +
<p>Furthermore, <span class="math-inline"><math>A</math></span> is said to be nowhere dense if the interior of the closure of <span class="math-inline"><math>A</math></span> is empty, that is:</p>
 +
<div style="text-align: center;"><math>\begin{align} \quad \mathrm{int} (\bar{A}) = \emptyset \end{align}</math></div>
 +
<p>We will now look at two very important definitions regarding whether an arbitrary set <span class="math-inline"><math>A \subseteq X</math></span> can either be written as the union of a countable collection of nowhere dense subsets of <span class="math-inline"><math>X</math></span> or not.</p>
 +
<blockquote style="background: white; border: 1px solid black; padding: 1em;">
 +
<td><strong>Definition:</strong> Let <span class="math-inline"><math>(X, \tau)</math></span> be a topological space. A set <span class="math-inline"><math>A \subseteq X</math></span> is said to be of <strong>The First Category</strong> or <strong>Meager</strong> if <span class="math-inline"><math>A</math></span> can be expressed as the union of a countable number of nowhere dense subsets of <span class="math-inline"><math>X</math></span>. If <span class="math-inline"><math>A</math></span> cannot be expressed as such a union, then <span class="math-inline"><math>A</math></span> is said to be of <strong>The Second Category</strong> or <strong>Nonmeager</strong>.</td>
 +
</blockquote>
 +
<p>Note that in general it is much easier to show that a set <span class="math-inline"><math>A \subseteq X</math></span> of a topological space <span class="math-inline"><math>(X, \tau)</math></span> is of the first category since we only need to find a countable collection of nowhere dense subsets, say <span class="math-inline"><math>\{ A_1, A_2, ... \}</math></span> (possibly finite) where each <span class="math-inline"><math>A_i</math></span> is nowhere dense such that:</p>
 +
<div style="text-align: center;"><math>\begin{align} \quad A = \bigcup_{i=1}^{\infty} A_i \end{align}</math></div>
 +
<p>Showing that <span class="math-inline"><math>A \subseteq X</math></span> is of the second category is much more difficult since we must show that no such union of a countable collection of nowhere dense subsets from <span class="math-inline"><math>X</math></span> equals <span class="math-inline"><math>A</math></span>.</p>
 +
<p>For an example of a set of the first category, consider the topological space <span class="math-inline"><math>(\mathbb{R}, \tau)</math></span> where <span class="math-inline"><math>\tau</math></span> is the usual topology of open intervals and consider the set <span class="math-inline"><math>\mathbb{Q} \subseteq \mathbb{R}</math></span> of rational numbers. We already know that the set of rational numbers is countable, so the following union is a union of a countable collection of subsets of <span class="math-inline"><math>X</math></span>:</p>
 +
<div style="text-align: center;"><math>\begin{align} \quad \mathbb{Q} = \bigcup_{q \in \mathbb{Q}} \{ q \} \end{align}</math></div>
 +
<p>Each of the sets <span class="math-inline"><math>\{ q \}</math></span> is nowhere dense. Therefore <span class="math-inline"><math>\mathbb{Q}</math></span> can be expressed as the union of a countable collection of nowhere dense subsets of <span class="math-inline"><math>X</math></span>, so <span class="math-inline"><math>\mathbb{Q}</math></span> is of the first category.</p>
 +
 
 +
==The Baire Category Theorem==
 
<blockquote style="background: white; border: 1px solid black; padding: 1em;">
 
<blockquote style="background: white; border: 1px solid black; padding: 1em;">
 
<td><strong>Lemma 1:</strong> Let <span class="math-inline"><math>(X, \tau)</math></span> be a topological space and let <span class="math-inline"><math>A \subseteq X</math></span>. If <span class="math-inline"><math>A</math></span> is a nowhere dense set then for every <span class="math-inline"><math>U \in \tau</math></span> there exists a <span class="math-inline"><math>B \subseteq U</math></span> such that <span class="math-inline"><math>A \cap \bar{B} = \emptyset</math></span>.</td>
 
<td><strong>Lemma 1:</strong> Let <span class="math-inline"><math>(X, \tau)</math></span> be a topological space and let <span class="math-inline"><math>A \subseteq X</math></span>. If <span class="math-inline"><math>A</math></span> is a nowhere dense set then for every <span class="math-inline"><math>U \in \tau</math></span> there exists a <span class="math-inline"><math>B \subseteq U</math></span> such that <span class="math-inline"><math>A \cap \bar{B} = \emptyset</math></span>.</td>
Line 25: Line 69:
 
<li>Since <span class="math-inline"><math>x \in \bar{B_n}</math></span> for all <span class="math-inline"><math>n \in \mathbb{N}</math></span> then since <span class="math-inline"><math>A_n \cap \bar{B_n} = \emptyset</math></span> we must have that then <span class="math-inline"><math> x \not\in A_n </math></span></li>
 
<li>Since <span class="math-inline"><math>x \in \bar{B_n}</math></span> for all <span class="math-inline"><math>n \in \mathbb{N}</math></span> then since <span class="math-inline"><math>A_n \cap \bar{B_n} = \emptyset</math></span> we must have that then <span class="math-inline"><math> x \not\in A_n </math></span></li>
 
</ul>
 
</ul>
 +
 +
==Licensing==
 +
Content obtained and/or adapted from:
 +
* [http://mathonline.wikidot.com/dense-and-nowhere-dense-sets-in-a-topological-space Dense and Nowhere Dense Sets in a Topological Space, mathonline.wikidot.com] under a CC BY-SA license
 +
* [http://mathonline.wikidot.com/sets-of-the-first-and-second-categories-in-a-topological-spa Sets of the First and Second Categories in a Topological Space, mathonline.wikidot.com] under a CC BY-SA license
 +
* [http://mathonline.wikidot.com/the-baire-category-theorem The Baire Category Theorem, mathonline.wikidot.com] under a CC BY-SA license

Latest revision as of 15:17, 8 November 2021

Dense and Nowhere Dense Sets

Dense Sets in a Topological Space

Definition: Let be a topological space. The set is said to be Dense in if the intersection of every nonempty open set with is nonempty, that is, for all .

Given any topological space it is important to note that is dense in because every is such that , and so for all .

For another example, consider the topological space where is the usual topology of open intervals. Then the set of rational numbers is dense in . If not, then there exists an such that .

Since we have that for some open interval with and . Suppose that . Then we must also have that:

The intersection above implies that there exists no rational numbers in the interval , i.e., there exists no such that . But this is a contradiction since for all with there ALWAYS exists a rational number such that , i.e., . So for all . Thus, is dense in .

We will now look at a very important theorem which will give us a way to determine whether a set is dense in or not.

Theorem 1: Let be a topological space and let . Then is dense in if and only if .

  • Proof: Suppose that is dense in . Then for all we have that . Clearly so we only need to show that .

Nowhere Dense Sets in a Topological Space

Definition: Let be a topological space. A set is said to be Nowhere Dense in if the interior of the closure of is empty, that is, .

For example, consider the topological space where is the usually topology of open intervals on , and consider the set of integers . The closure of , is the smallest closed set containing . The smallest closed set containing is since is open as is an arbitrary union of open sets:

So what is the interior of ? It is the largest open set contained in . All open sets of with respect to this topology are either the empty set, an open interval, a union of open intervals, or the whole set (the union of all open intervals). But no open intervals are contained in and so:

Therefore is a nowhere dense set in with respect to the usual topology on .


Sets of the First and Second Categories in a Topological Space

Recall that if is a topological space then a set is said to be dense in if the intersection of with all open sets (except for the empty set) is nonempty, that is, for all we have that:

Furthermore, is said to be nowhere dense if the interior of the closure of is empty, that is:

We will now look at two very important definitions regarding whether an arbitrary set can either be written as the union of a countable collection of nowhere dense subsets of or not.

Definition: Let be a topological space. A set is said to be of The First Category or Meager if can be expressed as the union of a countable number of nowhere dense subsets of . If cannot be expressed as such a union, then is said to be of The Second Category or Nonmeager.

Note that in general it is much easier to show that a set of a topological space is of the first category since we only need to find a countable collection of nowhere dense subsets, say (possibly finite) where each is nowhere dense such that:

Showing that is of the second category is much more difficult since we must show that no such union of a countable collection of nowhere dense subsets from equals .

For an example of a set of the first category, consider the topological space where is the usual topology of open intervals and consider the set of rational numbers. We already know that the set of rational numbers is countable, so the following union is a union of a countable collection of subsets of :

Each of the sets is nowhere dense. Therefore can be expressed as the union of a countable collection of nowhere dense subsets of , so is of the first category.

The Baire Category Theorem

Lemma 1: Let be a topological space and let . If is a nowhere dense set then for every there exists a such that .

Theorem 1 (The Baire Category Theorem): Every complete metric space is of the second category.

  • Proof: Let be a complete metric space. Then every Cauchy sequence of elements from converges in . Suppose that is of the first category. Then there exists a countable collection of nowhere dense sets such that:
  • Let . For each nowhere dense set , there exists a set such that .
  • Let be a ball contained in such that . Let be a ball contained in whose radius is and such that . Repeat this process. For each let be a ball contained in whose radius is and such that and such that .
  • The sequence is Cauchy since as gets large, the elements are very close. Since is a complete metric space, we must have that this Cauchy sequence therefore converges to some , i.e., .
  • Now notice that for all because if not, then there exists an such that for all . Hence is open and so there exists an open ball such that but then because for all .
  • Since for all then since we must have that then

Licensing

Content obtained and/or adapted from: