In mathematics, the linear algebra concept of dual basis can be applied in the context of a finite extensionL/K, by using the field trace. This requires the property that the field trace TrL/K provides a non-degenerate quadratic form over K. This can be guaranteed if the extension is separable; it is automatically true if K is a perfect field, and hence in the cases where K is finite, or of characteristic zero.
A dual basis isn't a concrete basis like the polynomial basis or the normal basis; rather it provides a way of using a second basis for computations.
Consider two bases for elements in a finite field, GF(pm):
and
then B2 can be considered a dual basis of B1 provided
Here the trace of a value in GF(pm) can be calculated as follows:
Using a dual basis can provide a way to easily communicate between devices that use different bases, rather than having to explicitly convert between bases using the change of bases formula. Furthermore, if a dual basis is implemented then conversion from an element in the original basis to the dual basis can be accomplished with a multiplication by the multiplicative identity (usually 1).
The transformation that describes the new basis vectors in terms of the old basis, is defined as a covariant transformation.
(the basis vectors are tangent vectors to the coordinate grid).
Since vectors (and dual vectors) are defined coordinate independently, this definition of a tensor is also free of coordinates and does not depend on the choice of a coordinate system.
In the language of category theory, taking the dual of vector spaces and the pullback of linear maps is therefore a contravariant functor from the category of vector spaces over F to itself.
The idea of a dual vector as an infinite sum should not be taken too literally; in general infinite sums are defined in terms of a limit, which only makes sense in a topological space, and even then not all sums will be convergent.
The continuous dual V′ of a normed vector space V (e.g., a Banach space or a Hilbert space) forms a normed vector space.