Skip to content

Group action, orbits, and stabilizers

Group action

Definition: If \(G\) is a group with identity element \(e\), and \(X\) is a set, then a group action \(\alpha\) of \(G\) on \(X\) is a function

\[ \alpha: G\times X\to X \]

that satisfies two axioms for all \(g,h\in G,x\in X\):

  • identity: \(\alpha(e,x)= x\)
  • compatibility \(\alpha(g,\alpha(h,x))=\alpha(gh,x)\)

with \(\alpha(g,x)\) often shortened to \(gx\) or \(g\cdot x\) when there's no ambiguity of the action being considered in the context. The group \(G\) is said to act on \(X\); a set \(X\) together with the action of \(G\) is called a \(G\)-set.

From these two axioms, we know that the function that maps \(x\) to \(g\cdot x\) is a bijection, since we have the inverse map \(g^{-1}\) such that \(g^{-1}\cdot (g\cdot x)=x\).

Orbits and stabilizers

Consider a group \(G\) acting on a set \(X\).

Definition: The orbit of an element \(x\in X\) is the set of elements in \(X\) which \(x\) can be moved to through the group action, denoted by \(G\cdot x\):

\[ G\cdot x=\{g\cdot x|g\in G\} \]

Proposition: If and only if there exists a \(g\in G\) such that \(g\cdot x =y\) for \(x,y\in X\), we say that \(x\sim y\). This is an equivalence relation.

Proof:

  • Reflectivity: $e\cdot x = x\implies x\sim x $
  • Symmetry: \(x\sim y\implies g\cdot x = y\implies x=g^{-1}\cdot y\implies y\sim x\)
  • Transitivity: \(x\sim y\land y\sim z\implies g\cdot x = y,h\cdot y=z\implies h\cdot(g\cdot x) =(hg)\cdot x=z\implies x\sim z\)

This proposition implies that the set of orbits of \(X\) under the group action \(G\) forms a partition of \(X\).

Definition: Given \(g\in G\) and \(x\in X\) with \(g\cdot x =x\), we say \(x\) is fixed under the action of \(g\). For every \(x\in X\), its stabilizer in \(G\) is the set of all elements \(g\in G\) that fixes \(x\), denoted as \(G_x\):

\[ G_x=\{g\in G|g\cdot x = x\} \]

The stabilizer is more often called a stabilizer subgroup! Indeed, we can prove this by verifying the conditions:

For \(g\in G_x\), we have

\[ x=e\cdot x=(g^{-1}g)\cdot x =g^{-1}\cdot(g\cdot x)=g^{-1}\cdot x \]

Thus \(g^{-1}\in G_x\).

The identity element \(e\) is clearly in \(G_x\) since \(e\cdot x =x\).

For \(g,h\in G_x\),

\[ gh\cdot x =g\cdot(h\cdot x)=g\cdot x =x\implies gh\in G_x \]

Thus, \(G_x\) is a subgroup of \(G\).

Orbit-stabilizer theorem

Theorem: For a finite group \(G\) acting on a set \(X\) and any element \(x\in X\).

\[ |G\cdot x| = [G:G_x] =\frac{|G|}{|G_x|} \]

Proof: For a fixed \(x\in X\), consider the map \(f:G\to X\) given by mapping \(g\) to \(g\cdot x\). By definition, the image of \(f(G)\) is the orbit of \(G\cdot x\). If two elements \(g,h\in G\) have the same image:

\[ f(g)=f(h)\Longleftrightarrow g\cdot x =h\cdot x\Longleftrightarrow g^{-1}h\cdot x = x\Longleftrightarrow g^{-1}h\in G_x\Longleftrightarrow h\in gG_x \]

In other words, \(f(g)=f(h)\) if and only if they are in the same coset for the stabilizer subgroup \(G_x\). Hence, \(f\) induces a bijection between the set \(G/G_x\) of cosets and the orbit \(G\cdot x\) which sends \(gG_x\mapsto g\cdot x\). Combining the bijection with lagrange theorem gives orbit-stabilizer theorem as desired.

Burnside's Lemma

This is a lemma useful in taking account of symmetry when counting mathematical objects. It can be proven by orbit-stabilizer theorem and we will show that in later sections.

Theorem: Let \(G\) be a finite group acting on a set \(X\). For each \(g\in G\), let \(X^g\) denote the set of elements in \(X\) that are fixed by \(g\):

\[ X^g=\{x\in X|g\cdot x = x\} \]

Burnside's lemma provides a formula for the number of orbits, denoted by \(X/G\):

\[ |X/G| = \frac{1}{|G|}\sum_{g\in G}|X^g| \]

Examples

First, we want to see how we can apply it in real life. Suppose we want to see how many ways we can color a cube with 3 colors, it would be simply \(3^6=729\) ways. However, what if two colorings are essentially the same after some rotations?

Let \(X\) be the set of all \(3^6\) color combination and \(G\) be the rotational symmtery group of the cube acting on \(X\) in the natural manner. There are \(6\times 4= 24\) elements in \(G\).

Then two elements of \(X\) belong to the same orbit if one can be rotated to the other. Hence, we are finding the number of orbits! Now, it suffices to calculate \(X^g\) for all \(g\in G\):

  • the identity element fixes all elements of \(X\).
  • six 90-degree face rotations, each fixes \(3^3\) elements of \(X\).
  • three 180-degree face rotations, each fixes \(3^4\) elements of \(X\).
  • eight vertex rotations, each fixes \(3^2\) elements of \(X\).
  • six edge flip, each fixes \(3^3\) elements of \(X\).

Hence, the average fix size is

\[ \frac1{24}(3^6+6\times 3^3+3\times 3^4+8\times 3^2+6\times 3^3)=57 \]

Finally, how do we prove this lemma?

Proof: First, we would like to re-express the sum over group elements \(g\in G\) as an equivalent sum over elements \(x\in X\):

\[ \sum_{g\in G}|X^g|=|\{(g,x)\in G\times X|g\cdot x = x\}|=\sum_{x\in X}|G_x| \]

By orbit-stabilizer theorem, we can rewrite the equation into:

\[ \sum_{x\in X}|G_x|=|G|\sum_{x\in X}\frac{1}{|G\cdot x|} \]

Finally, notice that the set of orbits partition \(X\), so we can break the sum into separate sums over each individual orbit:

\[ \sum_{x\in X}\frac{1}{|G\cdot x|}=\sum_{A\in X/G}\sum_{x\in A}\frac{1}{|A|}=|X/G| \]

Putting everything together, we get:

\[ |X/G| = \frac{1}{|G|}\sum_{g\in G}|X^g| \]

Comments