Angles between flats

Jordan's definition

Let F and G be flats of dimensions k and l in the n -dimensional Euclidean space E n . By definition, a translation of F or G does not alter their mutual angles. If F and G do not intersect, they will do so upon any translation of G which maps some point in G to some point in F . It can therefore be assumed without loss of generality that F and G intersect.

Jordan shows that Cartesian coordinates x 1 , … , x ρ , y 1 , … , y σ , z 1 , … , z τ , u 1 , … , u υ , v 1 , … , v α , w 1 , … , w α in E n can then be defined such that F and G are described, respectively, by the sets of equations

and

with 0 < θ i < π / 2 , i = 1 , … , α . Jordan calls these coordinates canonical. By definition, the angles θ i are the angles between F and G .

The non-negative integers ρ , σ , τ , υ , α are constrained by

For these equations to determine the five non-negative integers completely, besides the dimensions n , k and l and the number α of angles θ i , the non-negative integer σ must be given. This is the number of coordinates y i , whose corresponding axes are those lying entirely within both F and G . The integer σ is thus the dimension of F ∩ G . The set of angles θ i may be supplemented with σ angles 0 to indicate that F ∩ G has that dimension.

Jordan's proof applies essentially unaltered when E n is replaced with the n -dimensional inner product space C n over the complex numbers. (For angles between subspaces, the generalisation to C n is discussed by Galántai and Hegedũs in terms of the below variational characterisation.)

Angles between subspaces

Now let F and G be subspaces of the n -dimensional inner product space over the real or complex numbers. Geometrically, F and G are flats, so Jordan's definition of mutual angles applies. When for any canonical coordinate ξ the symbol ξ ^ denotes the unit vector of the ξ axis, the vectors y ^ 1 , … , y ^ σ , w ^ 1 , … , w ^ α , z ^ 1 , … , z ^ τ form an orthonormal basis for F and the vectors y ^ 1 , … , y ^ σ , w ^ 1 ′ , … , w ^ α ′ , u ^ 1 , … , u ^ υ form an orthonormal basis for G , where

Being related to canonical coordinates, these basic vectors may be called canonical.

When a i , i = 1 , … , k denote the canonical basic vectors for F and b i , i = 1 , … , l the canonical basic vectors for G then the inner product ⟨ a i , b j ⟩ vanishes for any pair of i and j except the following ones.

With the above ordering of the basic vectors, the matrix of the inner products ⟨ a i , b j ⟩ is thus diagonal. In other words, if ( a i ′ , i = 1 , … , k ) and ( b i ′ , i = 1 , … , l ) are arbitrary orthonormal bases in F and G then the real, orthogonal or unitary transformations from the basis ( a i ′ ) to the basis ( a i ) and from the basis ( b i ′ ) to the basis ( b i ) realise a singular value decomposition of the matrix of inner products ⟨ a i ′ , b j ′ ⟩ . The diagonal matrix elements ⟨ a i , b i ⟩ are the singular values of the latter matrix. By the uniqueness of the singular value decomposition, the vectors y ^ i are then unique up to a real, orthogonal or unitary transformation among them, and the vectors w ^ i and w ^ i ′ (and hence v ^ i ) are unique up to equal real, orthogonal or unitary transformations applied simultaneously to the sets of the vectors w ^ i associated with a common value of θ i and to the corresponding sets of vectors w ^ i ′ (and hence to the corresponding sets of v ^ i ).

A singular value 1 can be interpreted as cos 0 corresponding to the angles 0 introduced above and associated with F ∩ G and a singular value 0 can be interpreted as cos ⁡ π / 2 corresponding to right angles between the orthogonal spaces F ∩ G ⊥ and F ⊥ ∩ G , where superscript ⊥ denotes the orthogonal complement.

Variational characterisation

The variational characterisation of singular values and vectors implies as a special case a variational characterisation of the angles between subspaces and their associated canonical vectors. This characterisation includes the angles 0 and π / 2 introduced above and orders the angles by increasing value. It can be given the form of the below alternative definition. In this context, it is customary to talk of principal angles and vectors.

Definition

Let V be an inner product space. Given two subspaces U , W with dim ⁡ ( U ) = k ≤ dim ⁡ ( W ) := l , there exists then a sequence of k angles 0 ≤ θ 1 ≤ θ 2 ≤ ⋯ ≤ θ k ≤ π / 2 called the principal angles, the first one defined as

where ⟨ ⋅ , ⋅ ⟩ is the inner product and ∥ ⋅ ∥ the induced norm. The vectors u 1 and w 1 are the corresponding principal vectors.

The other principal angles and vectors are then defined recursively via

This means that the principal angles ( θ 1 , … , θ k ) form a set of minimized angles between the two subspaces, and the principal vectors in each subspace are orthogonal to each other.

Geometric example

Geometrically, subspaces are flats (points, lines, planes etc.) that include the origin, thus any two subspaces intersect at least in the origin. Two two-dimensional subspaces U and W generate a set of two angles. In a three-dimensional Euclidean space, the subspaces U and W are either identical, or their intersection forms a line. In the former case, both θ 1 = θ 2 = 0 . In the latter case, only θ 1 = 0 , where vectors u 1 and w 1 are on the line of the intersection U ∩ W and have the same direction. The angle θ 2 > 0 will be the angle between the subspaces U and W in the orthogonal complement to U ∩ W . Imagining the angle between two planes in 3D, one intuitively thinks of the largest angle, θ 2 > 0 .

Algebraic example

In 4-dimensional real coordinate space R⁴, let the two-dimensional subspace U be spanned by u 1 = ( 1 , 0 , 0 , 0 ) and u 2 = ( 0 , 1 , 0 , 0 ) , while the two-dimensional subspace W be spanned by w 1 = ( 1 , 0 , 0 , a ) / 1 + a 2 and w 2 = ( 0 , 1 , b , 0 ) / 1 + b 2 with some real a and b such that | a | < | b | . Then u 1 and w 1 are, in fact, the pair of principal vectors corresponding to the angle θ 1 with cos ⁡ ( θ 1 ) = 1 / 1 + a 2 , and u 2 and w 2 are the principal vectors corresponding to the angle θ 2 with cos ⁡ ( θ 2 ) = 1 / 1 + b 2

To construct a pair of subspaces with any given set of k angles θ 1 , … , θ k in a 2 k (or larger) dimensional Euclidean space, take a subspace U with an orthonormal basis ( e 1 , … , e k ) and complete it to an orthonormal basis ( e 1 , … , e n ) of the Euclidean space, where n ≥ 2 k . Then, an orthonormal basis of the other subspace W is, e.g.,

Basic properties

If the largest angle is zero, one subspace is a subset of the other.

If the smallest angle is zero, the subspaces intersect at least in a line.

The number of angles equal to zero is the dimension of the space where the two subspaces intersect.