Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Linear Help Asap

  1. Dec 6, 2004 #1
    ok guys.. i would really appreciate any help you may have..

    Anyone know how to prove the axioms of norms? there are 3 of them.
  2. jcsd
  3. Dec 6, 2004 #2


    User Avatar
    Science Advisor
    Homework Helper

    What properties of the norm exactly are you trying to prove? Many of the basic properties will follow directly from the definitions. Show what you've got so far and you'll have a better chance of getting help.
  4. Dec 7, 2004 #3

    James R

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    You can't prove an axiom, by definition.

    Perhaps you are asking how you can show that a particular example satisfies a set of axioms?
  5. Dec 7, 2004 #4
    See. that's what I thought too..that you can't prove it b/c it is true by definition., but my prof seems to want us to prove it with a given definition of the norm.

    he defines it to be , for a n x m matrix A, the norm of A=(sqrt(nm))(max{abs.val(aij)}), where the ij are subscripts.

    I guess he wants us to show how this particular definition of the norm is in fact a norm. Any suggestions?
  6. Dec 7, 2004 #5

    matt grime

    User Avatar
    Science Advisor
    Homework Helper

    Yes, just do it! Can you see why norm(a+b) <= norm(a) + norm(b)? It's a "just do it proof", in the sense of Polya.
  7. Dec 7, 2004 #6
    k here's some more im having trouble with..

    Prove the following test for convergence (using the above definition of the norm): If the sum, from k=0 to infinity, of the norm of Ak (where k is the subscript) converges then the sum of Ak) where k is the subscript) converges.

    And how do i use this to prove that the sum from k=0 to infinity of Ak/k! ( where k in the numerator is an exponent) always converges for any square matirs A?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook