# nLab representation sphere

Contents

### Context

#### Representation theory

representation theory

geometric representation theory

# Contents

## Idea

Given a linear representation of a group (topological group) $G$ in a vector space/Cartesian space $\mathbb{R}^n$, then the corresponding representation sphere is the one-point compactification of this $\mathbb{R}^n$ (the $n$-sphere) regarded as a G-space.

Representation spheres induce the looping and delooping which is used in RO(G)-graded equivariant cohomology theory, represented by genuine G-spectra in equivariant stable homotopy theory.

## Properties

### Equivariant stereographic projection

###### Proposition

(representation spheres of $V$ are unit spheres in $\mathbb{R} \oplus V$)

Let $G$ be a finite group and $V \in RO(G)$ a finite-dimensional linear representation of $G$.

Conside the unit sphere $S(\mathbb{R}\oplus V)$ where $\mathbb{R}$ carries the trivial representation. Then the stereographic projection homeomorphism

$S(\mathbb{R}\oplus V)\setminus \{(1,\mathbf{0})\} \stackrel{\simeq}{\longrightarrow} V$

is manifestly $G$-equivariant, with its inverse exhibiting $S(\mathbb{R}\oplus V)$ as the one-point compactification of $V$, hence

$S^V \simeq_G S(\mathbb{R}\oplus V) \,.$

This also shows that $S^V$ is a smooth manifold with smooth $G$-action.

(e.g. MP 04, p. 2)

### $G$-CW-Complex structure

###### Proposition

(G-representation spheres are G-CW-complexes)

For $G$ a compact Lie group (e.g. a finite group) and $V \in RO(G)$ a finite-dimensional orthogonal $G$-linear representation, the representation sphere $S^V$ admits the structure of a G-CW-complex.

###### Proof

Observe that we have a $G$-equivariant homeomorphism between the representation sphere of $V$ and the unit sphere in $\mathbb{R} \oplus V$, where $\mathbb{R}$ is the 1-dimensional trivial representation (Prop. )

(1)$S^V \;\simeq\; S(\mathbb{R} \oplus V) \,.$

It is thus sufficient to show that unit spheres in orthogonal representations admit G-CW-complex structure.

This in turn follows as soon as there is a $G$-equivariant triangulation of $S(\mathbb{R}\oplus V)$, hence a triangulation with the property that the $G$-action restricts to a bijection on its sets of $k$-dimensional cells, for each $k$. Because then if $G/H$ is an orbit of this $G$-action on the set of $k$-cells, we have a cell $G/H \times D^k$ of an induced G-CW-complex.

Since the unit spheres in (1) are smooth manifolds with smooth $G$-action, the existence of such $G$-equivariant triangulations follows for general compact Lie groups $G$ from the equivariant triangulation theorem (Illman 83).

More explicitly, in the case that $G$ is a finite group such an equivariant triangulation may be constructed as follows:

Let $\{b_1, b_2, \cdots, b_{n+1}\}$ be an orthonormal basis of $\mathbb{R} \oplus V$. Take then as vertices of the triangulation all the distinct points $\pm g(b_i) \in \mathbb{R} \oplus V$, and as edges the geodesics (great circle segments) between nearest neighbours of these points, etc.

## References

• Andrew Blumberg, Example 1.1.5 of Equivariant homotopy theory, 2017 (pdf, GitHub)

• Waclaw Marzantowicz, Carlos Prieto, The unstable equivariant fixed point index and the equivariant degree, Jourmal of the London Mathematical Society, Volume 69, Issue 1 February 2004 , pp. 214-230 (pdf, doi:10.1112/S0024610703004721)

Last revised on December 1, 2019 at 14:15:01. See the history of this page for a list of all contributions to it.