Cauchy-Schwarz Inequality Proof | MathWorld Demonstration and Solution

  • Thread starter Thread starter fluidistic
  • Start date Start date
  • Tags Tags
    Cauchy Proof
fluidistic
Gold Member
Messages
3,928
Reaction score
272

Homework Statement


I'm trying to follow the demonstration of the Cauchy-Schwarz's inequality proof given in http://mathworld.wolfram.com/SchwarzsInequality.html.
I am stuck at the last step, namely that \langle \bar g , f \rangle \langle f , \bar g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle \Rightarrow |\langle f , g \rangle |^2 \leq \langle f , f \rangle \langle g , g \rangle.


Homework Equations



I don't know.

The Attempt at a Solution


\langle \bar g , f \rangle \langle f , \bar g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle \Rightarrow \langle \bar f , g \rangle \langle \bar f , g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle. I'm stuck here.
I know that ||f||=\sqrt {\langle f , f \rangle} but I don't even know if I can use this fact. Any tip is appreciated.
 
Physics news on Phys.org
I've made some progress I think.
Mathworld didn't specify it explicitely but I think that f and g are real functions.
So that I reach \langle \bar g , f \rangle \langle f , \bar g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle \Rightarrow \langle \bar f , g \rangle \langle \bar f , g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle \Rightarrow \langle f , g \rangle ^2 \leq \langle f , f \rangle \langle g , g \rangle. So I "almost" reach the proof. I have a missing absolute value though. Any idea why?
 
Ok I got it wrong, f and g aren't real valued function because lambda (which is complex) is defined by some inner products involving f and g and their complex conjugate only.
If someone could tell me how to understand the last step I'd be grateful.
 
fluidistic said:
I've made some progress I think.
Mathworld didn't specify it explicitely but I think that f and g are real functions.
So that I reach \langle \bar g , f \rangle \langle f , \bar g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle \Rightarrow \langle \bar f , g \rangle \langle \bar f , g \rangle \leq \langle \bar f , f \rangle \langle \bar g , g \rangle \Rightarrow \langle f , g \rangle ^2
This is incorrect. \langle f, g\rangle^2 is a complex number. You want \left|\langle f, g\rangle\right|^2.

\leq \langle f , f \rangle \langle g , g \rangle. So I "almost" reach the proof. I have a missing absolute value though. Any idea why?
Again, those are wrong. You want \langle f, \bar f\rangle= |f|^2 and \langle g, \bar g\rangle= |g|^2.
 
Thanks HallsofIvy!
HallsofIvy said:
This is incorrect. \langle f, g\rangle^2 is a complex number. You want \left|\langle f, g\rangle\right|^2.
Yeah you are right, I realized this in my previous post.


You want \langle f, \bar f\rangle= |f|^2 and \langle g, \bar g\rangle= |g|^2.
Ok... How do I prove these, for any inner product? I'm looking at the properties of Hermitian inner product given there: http://mathworld.wolfram.com/HermitianInnerProduct.html but I've no clue how to relate it with the |.|^2 (same as norm squared?)
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top