Skip to main content
\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)
Mathematics LibreTexts

9.2 Norms


\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)

The norm of a vector in an arbitrary inner product space is the analog of the length or magnitude of a vector in \(\mathbb{R}^{n}\). We formally define this concept as follows.

Definition 9.2.1.  Let \(V \) be a vector space over \(\mathbb{F}\). A map
    \norm{\cdot} : V &\to \mathbb{R}\\
                                  v &\mapsto \norm{v}
is a norm on \(V \) if the following three conditions are satisfied.

  1. Positive definiteness: \(\norm{v}=0 \) if and only if \(v=0\);
  2. Positive Homogeneity: \(\norm{av}=|a|\,\norm{v} \) for all \(a\in \mathbb{F} \) and \(v\in V\);
  3. Triangle inequality: \(\norm{v+w}\le \norm{v}+\norm{w} \) for all \(v,w\in V\).

Remark 9.2.2.  Note that, in fact, \(\norm{v}\ge 0 \) for each \(v\in V \) since

\[  0 = \norm{v-v} \le \norm{v} + \norm{-v} = 2\norm{v}. \]

Next we want to show that a norm can always be defined from an inner product \(\inner{\cdot}{\cdot} \) via the formula

\[   \norm{v} = \sqrt{\inner{v}{v}} ~\text{for all}   ~ v \in V  .  \tag{9.2.1} \] 

Properties 1 and 2 follow easily from Conditions~1 and 3 of Definition~9.1.1. The triangle inequality requires more careful proof, though, which we give in Theorem~9.3.4\ref{thm:triangle} in the next chapter.

       If we take \(V=\mathbb{R}^n\), then the norm defined by the usual dot product is related to the usual notion of length of a vector. Namely, for \(v=(x_1,\ldots,x_n)\in \mathbb{R}^n\), we have                                                                                      
\begin{equation} \label{eqn:NormInRn}
    \norm{v} = \sqrt{x_1^2+\cdots + x_n^2}.      \tag{9.2.2}

We illustrate this for the case of \(\mathbb{R^3} \) in Figure 9.2.1.




Figure 9.2.1: The length of a vector in \( \mathbb{R^3} \) via equation 9.2.1.


      While it is always possible to start with an inner product and use it to define a norm, the converse requires more care.  In particular, one can prove that a norm can be used to define an inner product via Equation 9.2.1 if and only if the norm satisfies the Parallelogram Law (Theorem 9.3.6~\ref{thm:ParallelogramLaw}).