What are the conditions for commuting linear maps?

  • Context: Undergrad 
  • Thread starter Thread starter Marin
  • Start date Start date
  • Tags Tags
    Linear
Click For Summary

Discussion Overview

The discussion revolves around the conditions under which two linear maps commute. Participants explore both specific cases involving matrices and more general conditions applicable to linear maps without matrix representation. The conversation also touches on the implications of changing bases on the commutation of linear maps.

Discussion Character

  • Exploratory
  • Technical explanation
  • Conceptual clarification
  • Debate/contested

Main Points Raised

  • One participant suggests that for two linear maps to commute, the condition AB - BA = 0 must hold when represented by matrices A and B.
  • Another participant clarifies that for two linear maps f and g to commute, the composition must satisfy f ∘ g = g ∘ f, which implies that the maps must be defined on the same vector spaces.
  • It is noted that if the linear maps are endomorphisms (maps from a vector space to itself), then the matrices representing them must be square.
  • One participant points out that being square is not sufficient for commutation, as there are square matrices that do not commute.
  • A later reply reiterates that the condition for commutation is that f ∘ g = g ∘ f, which translates to AB = BA for matrices, suggesting no simpler equivalent conditions exist.
  • Another participant proposes that two linear maps commute if they commute on all vectors of some basis, and explains that this property is preserved under a change of basis.

Areas of Agreement / Disagreement

Participants generally agree on the necessity of the condition f ∘ g = g ∘ f for commutation, but there is no consensus on whether additional conditions are required beyond the maps being endomorphisms and square matrices. The discussion remains unresolved regarding the sufficiency of these conditions.

Contextual Notes

Some limitations include the dependence on the definitions of linear maps and the specific vector spaces involved. The discussion does not resolve whether there are additional necessary conditions for commutation beyond those mentioned.

Marin
Messages
192
Reaction score
0
Hi all!

I was wondering what conditions must two linear maps obey in order to commute?

If they are described by two matrices A and B, then one condition would be:

AB-BA=0

but what if we don't know the matrices, so we cannot compute AB adn BA? How is one supposed to proceed, is there a more general condition?



And another question: Suppose A and B commute. Will they still commute if I change the basis, I mean is the commutation coordinate independent?


thanks a lot for the help,

marin
 
Physics news on Phys.org
I hope this answers in part your questions:

If you have two linear maps
f: V \rightarrow W \mbox{ and } g: V' \rightarrow W'
if you want the two maps to commute then what you are saying is you want the composition of maps to satisfy
f \circ g = g \circ f \hspace{0.3cm} \Rightarrow f(g(v)) = g(f(v)).
If this is the case then you need that
V = W = V' = W'
for the composition to make sense. In terms of matrices this implies that they are square.

In response to your question about changing the basis as
V = W = V' = W'
you must perform the basis transformation on each space and if this happens then commuting matrices will still commute under a change of basis.
 
ok, so I need my linear maps to be endomorphisms and the matrices to be square :)

but this is not enough, since there are plenty of square matrices which do not commute.

So what else do I need?
 
The only thing you will need is
f \circ g = g \circ f \hspace{0.3cm} \Rightarrow f(g(v)) = g(f(v)).

If the linear maps are matrices, A and B, then \circ is just the usual matrix multiplication which gives
AB = BA.
This is exactly what you had. I don't think that there is anything else needed and I can't think of any simpler equivalent conditions.
 
One condition is that two linear maps commute if and only if they commute on all vectors of some basis. To see that commutation is independent of basis, note that in a new basis, the transformation A becomes PAP^{-1} and B becomes PBP^{-1} for some change of basis matrix P. If A and B commute then (PAP^{-1})(PBP^{-1}) = PABP^{-1} = PBAP^{-1} = (PBP^{-1})(PAP^{-1}).
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
8K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 26 ·
Replies
26
Views
6K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 19 ·
Replies
19
Views
4K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 3 ·
Replies
3
Views
1K