PlanetMath: face of a convex set

(more info)

Math for the people, by the people.

Main Menu

sections Encyclopædia
Papers
Books
Expositions

meta Requests (198)
Orphanage
Unclass'd
Unproven (490)
Corrections (7)
Classification

talkback Polls
Forums
Feedback
Bug Reports

downloads Snapshots
PM Book

information News
Docs
Wiki
ChangeLog
TODO List
Legalese
About

face of a convex set

(Definition)

Let be a convex set in $\mathbb{R}^n$ (or any topological vector space). A face of is a subset of such that

is convex, and
given any line segment $L\subseteq C$ , if $\operatorname{ri}(L)\cap F\ne \varnothing$ , then $L\subseteq F$ .

Here, $\operatorname{ri}(L)$ denotes the relative interior of

(open segment of

A zero-dimensional face of a convex set is called an extreme point of .

This definition formalizes the notion of a face of a convex polygon or a convex polytope and generalizes it to an arbitrary convex set. For example, any point on the boundary of a closed unit disk in $\mathbb{R}^2$ is its face (and an extreme point).

Observe that the empty set and itself are faces of . These faces are sometimes called improper faces, while other faces are called proper faces.

Remarks. Let be a convex set.

The intersection of two faces of is a face of .
A face of a face of is a face of .
Any proper face of lies on its relative boundary, $\operatorname{rbd}(C)$ .
The set $\operatorname{Part}(C)$ of all relative interiors of the faces of partitions .
If is compact, then is the convex hull of its extreme points.
The set of faces of a convex set forms a lattice, where the meet is the intersection: $F_1 \wedge F_2 := F_1\cap F_2$ ; the join of is the smallest face $F\in F(C)$ containing both and . This lattice is bounded lattice (by $\varnothing$ and ). And it is not hard to see that is a complete lattice.
However, in general, is not a modular lattice. As a counterexample, consider the unit square $[0,1]\times [0,1]$ and faces , $b=\lbrace (0,y)\mid y\in [0,1]\rbrace$ , and . We have $a\le b$ . However, $a\vee (b\wedge c)=(0,0)\vee \varnothing=(0,0)$ , whereas $(a\vee b)\wedge c = b\wedge \varnothing=\varnothing$ .
Nevertheless, is a complemented lattice. Pick any face $F\in F(C)$ . If , then $\varnothing$ is a complement of . Otherwise, form $\operatorname{Part}(C)$ and $\operatorname{Part}(F)$ , the partitions of and into disjoint unions of the relative interiors of their corresponding faces. Clearly $\operatorname{Part}(F)\subset \operatorname{Part}(C)$ strictly. Now, it is possible to find an extreme point such that $\lbrace p\rbrace\in \operatorname{Part}(C)-\operatorname{Part}(F)$ . Otherwise, all extreme points lie in $\operatorname{Part}(F)$ , which leads to
$\displaystyle \operatorname{Part}(F) = \operatorname{Part}($ convex hull of extreme points of $\displaystyle C)=\operatorname{Part}(C),$
a contradiction. Finally, let be the convex hull of extreme points of not contained in $\operatorname{Part}(F)$ . We assert that is a complement of . If $x\in G\cap F$ , then $G\cap F$ is a proper face of and of , hence its extreme points are also extreme points of , and of , which is impossible by the construction of . Therefore $F\cap G=\varnothing$ . Next, note that the union of extreme points of and of is the collection of all extreme points of , this is again the result of the construction of , so any $y\in C$ is in the join of all its extreme points, which is equal to the join of and (since join is universally associative).
Additionally, in , zero-dimensional faces are compact elements, and compact elements are faces with finitely many extreme points. The unit disk is not compact in . Since every face is the convex hull (join) of all extreme points it contains, is an algebraic lattice.