Compact space - Properties and Advanced Topics

Properties of Compact Spaces Introduction Compact spaces are among the most important objects in topology. Their value comes from the remarkable collection of properties they possess—properties that often generalize familiar facts from the real numbers to much more abstract settings. In this section, we explore the fundamental characteristics of compact spaces and how they interact with continuous functions and other spaces. Separation and Closure Properties One of the most powerful results connecting compactness to separation is the relationship between compact sets and closed sets. In a Hausdorff space, every compact subset is closed. This is a critical theorem. To understand why this matters, recall that closed sets are often "easier" to work with than arbitrary sets, so knowing a set is closed gives us significant structural information. The key intuition is that Hausdorff spaces have enough separation (any two distinct points can be separated by disjoint open sets) that when combined with compactness, we can prove any point outside a compact set can be isolated from it, making the complement open and the set itself closed. However, this result depends fundamentally on the Hausdorff assumption. In a non-Hausdorff space, a compact subset need not be closed, and the closure of a compact set may fail to be compact. This shows that the strength of a topology matters—weaker topologies can allow pathological behavior. Understanding this limitation is important: compactness alone doesn't guarantee closure without additional structure. Normality and Regularity Beyond closure, compact Hausdorff spaces enjoy even stronger separation properties. Every compact Hausdorff space is normal and regular. Regularity means that a point and a closed set not containing it can be separated by disjoint open neighborhoods. Normality means that any two disjoint closed sets can be separated by disjoint open neighborhoods. These are powerful separation axioms that guarantee the space is "well-behaved" in terms of how points and sets relate to each other. For compact Hausdorff spaces, these properties come for free—compactness and the Hausdorff property together force the space to satisfy these strong separation conditions. Separation of Disjoint Compact Sets A natural question arises: can we separate disjoint compact sets the way we separate disjoint closed sets? The answer is yes, and this is a beautiful result. In a Hausdorff space, any two disjoint compact subsets can be separated by disjoint open neighborhoods. This means if $K1$ and $K2$ are compact and disjoint, there exist open sets $U$ and $V$ with $K1 \subseteq U$, $K2 \subseteq V$, and $U \cap V = \emptyset$. The image illustrates this concept: sets A and B (or B and C) are disjoint compact subsets that can each be enclosed in disjoint open neighborhoods. This property is particularly useful because it applies in Hausdorff spaces without requiring compactness of the entire space—only the subsets we're trying to separate need to be compact. Maps and Continuous Functions Homeomorphisms from Compact Spaces One of the most elegant theorems in topology concerns continuous bijections from compact spaces: A continuous bijection from a compact space onto a Hausdorff space is a homeomorphism. This is remarkably powerful. Normally, to prove a function is a homeomorphism, we must show both the function and its inverse are continuous. But if the domain is compact and the codomain is Hausdorff, the continuity of the original function automatically forces the inverse to be continuous. Intuitively, this is because the compact space and the structure of open and closed sets interact in a way that makes it impossible for a continuous bijection to have a discontinuous inverse. The Extreme Value Theorem The extreme value theorem is perhaps the most practically important theorem about continuous functions on compact spaces: A continuous real-valued function on a non-empty compact space attains its maximum and minimum and is bounded. More formally, if $f: K \to \mathbb{R}$ is continuous and $K$ is a non-empty compact space, then there exist points $x{\max}, x{\min} \in K$ such that $$f(x{\min}) \leq f(x) \leq f(x{\max})$$ for all $x \in K$. This generalizes the familiar result from calculus that a continuous function on a closed and bounded interval $[a,b]$ attains its maximum and minimum. The abstract version shows that it's really compactness doing the work, not the specific structure of the real line. Uniform Convergence Another important result concerns sequences of continuous functions: The uniform limit of a sequence of continuous functions on a compact space is continuous. This means if $fn: K \to \mathbb{R}$ is a sequence of continuous functions on a compact space $K$, and if $fn$ converges uniformly to some function $f$, then $f$ is also continuous. (Compare this to pointwise convergence, where the limit can be discontinuous.) The compactness of the domain is essential here—the same result fails on non-compact spaces without additional conditions. Boundedness in Metric Spaces When working in metric spaces, compactness has a concrete consequence: In any metric space, a compact subset is bounded. A set is bounded if its diameter is finite—that is, if the maximum distance between any two points in the set is finite. For compact subsets of metric spaces like $\mathbb{R}^n$, this means they fit inside a ball of some finite radius. This is intuitive: a compact set cannot "spread out infinitely" in a metric space. Note: boundedness in metric spaces is a different notion from compactness (in fact, $\mathbb{R}$ is non-compact but we can find closed bounded sets like $[-1,1]$ that are compact). But the direction from compactness to boundedness is always true. Compactifications Sometimes we want to make a non-compact space into a compact space by adding points. This process is called compactification. Every topological space can be embedded as an open dense subspace of a compact space by adding at most one extra point (Alexandroff one-point compactification). The Alexandroff compactification is constructed by taking a space $X$ and adding a single "point at infinity," denoted $\infty$. The topology is defined so that: Every open set of $X$ remains open in the compactification Open sets containing $\infty$ are complements of compact closed sets in $X$ The result is a compact space containing $X$ as a dense subset. Every locally compact Hausdorff space has a one-point compactification that is also Hausdorff. This is particularly nice because it preserves the Hausdorff property—a strong condition ensuring good separation. <extrainfo> A practical note: not all one-point compactifications are Hausdorff if the original space isn't locally compact. For instance, the one-point compactification of the rationals $\mathbb{Q}$ (which isn't locally compact) is not Hausdorff. </extrainfo> Ordered Compact Spaces When we have a total order structure, compactness has an immediate consequence: Any non-empty compact subset of the real numbers possesses a greatest element and a least element. This is a consequence of the extreme value theorem: if $K$ is a compact non-empty subset of $\mathbb{R}$, then the identity function (which is continuous) attains its maximum and minimum on $K$. This means there are points $\max(K)$ and $\min(K)$ in $K$. This illustrates a general principle: when compactness combines with additional structure (like an ordering), it produces concrete and useful results. <extrainfo> Algebraic Examples: Topological Groups A final note on concrete examples: compact groups include orthogonal groups (like the rotation group $SO(n)$), while non-compact groups include general linear groups (like $GL(n, \mathbb{R})$). These are examples of how the abstract properties we've discussed apply to important mathematical objects. The compactness of orthogonal groups versus the non-compactness of general linear groups has profound consequences for representation theory and harmonic analysis, though those applications are beyond the current scope. </extrainfo>