Confusion between orthogonal sum and orthogonal direct sum

Click For Summary
SUMMARY

The discussion clarifies the distinction between orthogonal direct sums and orthogonal sums in vector spaces. An orthogonal direct sum, denoted by the symbol \oplus, can represent both a Cartesian product of vector spaces and a direct sum of a subspace with its orthogonal complement. The orthogonal decomposition theorem states that a vector space can be expressed as an internal direct sum of a subspace and its orthogonal complement, specifically in the context of inner product spaces. The conversation emphasizes the importance of context when using the term "orthogonal decomposition" and the necessity of distinguishing between internal and external direct sums.

PREREQUISITES
  • Understanding of vector spaces and their properties
  • Familiarity with inner product spaces
  • Knowledge of direct sums and orthogonal complements
  • Basic concepts of linear algebra and notation
NEXT STEPS
  • Study the Orthogonal Decomposition Theorem in detail
  • Learn about internal and external direct sums in vector spaces
  • Explore the properties of inner product spaces and their applications
  • Review examples of orthogonal sums in linear algebra
USEFUL FOR

Mathematics students, educators, and professionals in linear algebra, particularly those focusing on vector space theory and inner product spaces.

logarithmic
Messages
103
Reaction score
0
For 2 vector spaces an orthogonal direct sum is a cartesian product of the spaces (with some other stuff) (http://planetmath.org/encyclopedia/OrthogonalSum.html ), and this orthogonal direct sum uses the symbol, \oplus.

However, there's an orthogonal decomposition theorem (http://planetmath.org/encyclopedia/OrthogonalDecompositionTheorem.html ), which says a vector space can be written as the direct sum of a subspace and it's orthogonal complement. But this direct sum, also denoted by the symbol \oplus, is not the cartesian product of the 2 spaces.

So A\oplus B, sometimes is the Cartesian product of A and B (the first paragraph), and other times it's \{a+b|a\in A, b \in B\} (the second paragraph).

I have a book which mentions a space X being decomposed into an "orthogonal sum of subspaces", and writes: X=\oplus_{i=1}^\infty Y_i. I think the \oplus here means the definition in the second paragraph is this correct?

Can someone check that my understanding of the difference between an orthogonal direct sum and an orthogonal sum, and my use of terminology in this post is correct.
 
Last edited by a moderator:
Physics news on Phys.org
I think what's confusing you is the difference between an "internal direct sum" (in which you write a vector space as a direct sum of subspaces) and an "external direct sum" (in which you take two vector spaces and add them together, without reference to them being contained in a larger vector space).

Orthogonal decomposition should (in my opinion) always be reserved for the situation where there is an inner product (or something similar) available, so that we can actually talk about things being orthogonal.

Anyway, note that if V is an inner product space and U is a subspace of V then the "orthogonal decomposition" ##V=U\oplus U^\perp## is in fact an internal direct sum.

To answer your question about the terminology in your book, it would help if you told us what the book is, or provided us with more context.
 
If an inner product space X is an internal orthogonal direct sum of two of its subspaces Y and Z, then X is isomorphic to the external orthogonal direct sum of Y and Z.

The former statement means that X has subspaces Y and Z such that Y is orthogonal to Z, and for each x in X, there's a unique pair (y,z) such that y is in Y, z is in Z, and x=y+z. The external orthogonal direct sum of two inner product spaces Y and Z is defined as the the set of ordered pairs (y,z) such that y is in Y and z is in Z, with addition and scalar multiplication defined in the obvious ways, and the inner product defined by ##\langle (y_1,z_1),(y_2,z_2)\rangle=\langle y_1,y_2\rangle_Y+\langle z_1,z_2\rangle_Z##.

If you now define Y' as the subspace that consists of ordered pairs (y,0), and define Z' similarly, then the external orthogonal direct sum of Y and Z is an internal orthogonal direct sum of Y' and Z'.

I think the things I have mentioned in here are the reasons why authors sometimes don't even bother to mention if what they have in mind is an internal orthogonal direct sum or an external orthogonal direct sum.
 
Last edited:

Similar threads

  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 3 ·
Replies
3
Views
5K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
6
Views
4K