so norm of a number gives out the length

norm of a vector gives out the magnitude

but what about p norm?

what is a p norm and what does it solve for?

Printable View

- Dec 10th 2009, 01:23 AMbnr34rb26dettWhat is a Norm? Norm of a vector? P-Norm?
so norm of a number gives out the length

norm of a vector gives out the magnitude

but what about p norm?

what is a p norm and what does it solve for? - Dec 10th 2009, 02:06 AMcraig
This link may be of help to you

Matrix Norm -- from Wolfram MathWorld - Dec 10th 2009, 03:22 AMCaptainBlack
- Dec 10th 2009, 06:06 AMHallsofIvy
In general, a "norm" on a vector space is a function that assigns to each vector a number, ||v||, satifying:

1) ||u+ v|| ||u||+ ||v||.

2) ||v||= 0 if and only if v is the 0 vector.

3) ||av||= |a|||v|| for any scalar (number) a, and vector v.

The formula Captain Black gives defines the "p" norm, over , for any positive number p (p is usually taken to be an integer but that is not necessary)- though I believe it should include an absolute value:

.

Of course, the p-norm for p= 2 is the usual "Euclidean norm" .

If p= 1 we get the "one-norm" .

Sometimes, although it doesn't fit the formula above, we define the "0-norm" to be - that is, take the absolute value of all components, then select the largest to be the norm.