I have been given the definition of a subordinate (operator or matrix) norm: $$\lvert\lvert A \rvert\rvert=\sup_{x\in V}\frac{\lvert\lvert Ax \rvert\rvert}{\lvert\lvert x \rvert\rvert},$$ where $V$ is a vector space, and $A$ is a linear operator. I have been told that in the case of a finite-dimensional vector space, we can replace $\sup$ with $\max$, giving $$\lvert\lvert A \rvert\rvert=\max_{x\in V}\frac{\lvert\lvert Ax \rvert\rvert}{\lvert\lvert x \rvert\rvert},$$ and I am wondering why this is true. It seems reasonable to me, but how can I be convinced that it's true?


First, see that you only need to consider vectors of norm $1$,

$$\lVert A\rVert = \sup_{\lVert x\rVert = 1} \lVert Ax\rVert.$$

Then note that the unit sphere in a finite-dimensional space is compact, and the function $x\mapsto \lVert Ax\rVert$ is continuous. A continuous real-valued function attains its maximum on a compact set, hence the supremum is a maximum in that case.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.