I am new to the concept of Dual space. Can someone please explain which of these are dual spaces and why? Also if $y$ is a polynomial on $x$, then what does $x(0)$ means?

• For $x \in P$, $y(x) = x(0)$

• Any vector space V, $y(x) = 0$; for every $x \in V $


As Pedro said in his answer, the dual space of a vector space $V$ is a set whose elements are linear functions from $V$ to what Pedro called the "underlying field". If this is unfamiliar to you, that probably means this mysterious "underlying field" is $\mathbb{R}$, the real numbers. So the dual space (sometimes denoted $V^*$) is the set of all linear functionals, i.e. all linear maps $$T: V \to \mathbb{R}$$

In other words, once you pick a basis for $V$, the dual space is just all functions represented by a row matrix $$(a_1 \ a_2 \ \dots \ a_n)$$ For example, if $V$ is the set of all $2 \times 2$ matrices, then the trace map $Tr: V \to \mathbb{R}$ is a linear functional (ie a linear map from $V$ to $\mathbb{R}$), since $$Tr(xA + yB) = x \ Tr(A) + y \ Tr(B)$$ where $x$ and $y$ are real numbers ("scalars") and $A$ and $B$ are matrices. Mind you, this is not the entire dual space! This is just one element of the dual space. So $Tr \in V^*$. If you pick the basis $$\{ \begin{pmatrix}1 & 0 \\ 0 & 0 \end{pmatrix} \begin{pmatrix}0 & 1 \\ 0 & 0 \end{pmatrix} \begin{pmatrix}0 & 0 \\ 1 & 0 \end{pmatrix} \begin{pmatrix}0 & 0 \\ 0 & 1 \end{pmatrix}\}$$ for $V$, then the map $Tr$ is given by the row matrix $$(1 \ 0 \ 0 \ 1)$$

Footnote: The reason for saying "underlying field" is that linear algebra can be done over many different "objects": the rational numbers, complex numbers, and finite fields - integers modulo a prime number $p$, and oftentimes even the integers or other "rings". But if that's not a point of interest now, then that's fine.

  • $\begingroup$ thanks for explaining so well, can you give a similar example for the polynomial problem above too? $\endgroup$ – Manish Aug 3 '14 at 6:29
  • 2
    $\begingroup$ @Manish Why don't you try working it out, and I'll let you know how it looks? (As many have said, "Math is not a spectator sport.") $\endgroup$ – Andrew Maurer Aug 3 '14 at 7:23

You're thinking about a linear mapping $V\to k$ where $k$ is the underlying field of your vector space. This is usually called a linear functional. These are the elements of the dual space of a vector space $V$, the linear mappings $V\to k$ (so the dual space of a vector space $V$ is not a mapping, it is the set of all linear mapings $V\to k$). As an example, you have above the linear functional that takes an element in $V=k[X]$, a polynomial with coefficients in your field, and returns the value of that polynomial at $0$ -- this is usually called an evaluation map (for obvious reasons), and if we're evaluating at $t\in k$; we usually denote it by ${\rm ev}_t$. So ${\rm ev}_t(p)=p(t)$.

You can indeed check that $p\mapsto p(0)$ where $p$ is a polynomial is linear, and $p(0)\in k$, so you have an element of $V^\ast:={\rm Hom}_k(V,k)$ (that's the notation for the dual space of $V$). The second example is the zero map from $V\to k$; it sends every element of $V$ to $0\in k$ (the additive identity of your field.) This is again linear, so it is an element of $V^\ast$.

  • $\begingroup$ can u please give me a simple example of both of them. Prefarably along with explanations. $\endgroup$ – Manish Aug 3 '14 at 5:12
  • $\begingroup$ I just did that. $\endgroup$ – Pedro Tamaroff Aug 3 '14 at 5:12

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.