Compactness in Metric Spaces
Compactness generalizes the Heine-Borel property to arbitrary metric spaces. A space is compact if every open cover has a finite subcover (or equivalently, every sequence has a convergent subsequence). Compact spaces share many properties with finite sets: continuous functions attain extrema, and spaces are complete and totally bounded.
Definition
A metric space is compact if every open cover has a finite subcover. Equivalently, is sequentially compact: every sequence in has a subsequence converging to a point in .
In metric spaces (unlike general topological spaces), compactness and sequential compactness are equivalent. This is not true in arbitrary topological spaces.
By the Heine-Borel theorem, is compact. More generally, closed bounded subsets of are compact.
is not compact. The sequence has no convergent subsequence in (the limit ).
The unit sphere is compact (closed and bounded in ).
Properties of compact spaces
Every compact metric space is complete and totally bounded.
If is compact and is continuous, then is compact.
As in , continuous functions on compact spaces attain their maximum and minimum.
Compactness in function spaces
In with the sup norm, a set is compact iff it is closed, bounded, and equicontinuous (Arzelà-Ascoli theorem). This is not the same as "closed and bounded" — boundedness alone is insufficient.
Summary
Compactness in metric spaces:
- Equivalent to sequential compactness (every sequence has convergent subsequence).
- Compact spaces are complete and totally bounded.
- Continuous images are compact (EVT holds).
- In : compact closed and bounded (Heine-Borel).
See Heine-Borel and Arzelà-Ascoli.