Prove Rk(A+B) ≤ Rk(A) + Rk(B) - Tal

  • Thread starter Thread starter talolard
  • Start date Start date
  • Tags Tags
    Matrix rank
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
1 reply · 2K views
talolard
Messages
119
Reaction score
0
Hey Guys, Another matrice question

Homework Statement


Prove: Rk(A+B)[tex]\leq[/tex] Rk(A) +Rk(B)



The Attempt at a Solution



Rk(A+B) = Dim[R(A) + R(B)]
Where R(A) is the row space of A
we know that Dim[R(A)+R(B)] = Dim[R(A)] + Dim[R(B)] - Dim[R(A)[tex]\cap[/tex]R(B)]
Which means that Dim[R(A)+R(B)] [tex]\leq[/tex] Dim[R(A)] + Dim[R(B)] iff Rk(A+B)[tex]\leq[/tex] Rk(A) +Rk(B)

I heard a rumor that this can also be done with linear transformations, can anyone elighten me on that path?

Is this correct?
Thanks
Tal
 
Physics news on Phys.org
talolard said:
Hey Guys, Another matrice question

Homework Statement


Prove: Rk(A+B)[tex]\leq[/tex] Rk(A) +Rk(B)



The Attempt at a Solution



Rk(A+B) = Dim[R(A) + R(B)]
Where R(A) is the row space of A
we know that Dim[R(A)+R(B)] = Dim[R(A)] + Dim[R(B)] - Dim[R(A)[tex]\cap[/tex]R(B)]
Which means that Dim[R(A)+R(B)] [tex]\leq[/tex] Dim[R(A)] + Dim[R(B)] iff Rk(A+B)[tex]\leq[/tex] Rk(A) +Rk(B)

I heard a rumor that this can also be done with linear transformations, can anyone elighten me on that path?
If F is a linear transformation from U to V, then, given specific bases for U and V, there exist a matrix representing F so essentially we can interpret matrices as being linear transformations and vice versa. Any thing true of matrices is true of linear transformations.

Is this correct?
Thanks
Tal