196 Type-(q,r) Tensor
A multilinear mapping that takes a fixed number of covectors and vectors to a scalar that is used to classify tensors by index type.
definition [d] (Type-\((q,r)\) Tensor = Type-\((r,s)\) Tensor = Rank \((r,s)\)) A multilinear map taking \(r\) covectors and \(s\) vectors to a scalar,
- \(T^{r}_{\ s} : \underbrace{V^{*}\times\cdots\times V^{*}}_{r} \times \underbrace{V\times\cdots\times V}_{s} \rightarrow \mathbb{R}\) .
where
- \(T^{r}_{\ s}\) is a tensor of type \((r,s)\).
- \(V\) is a vector space.
- \(V^{*}\) is the dual space of \(V\).
- \(r\) is the contravariant degree.
- \(s\) is the covariant degree.
- \(\mathbb{R}\) is the set of real numbers.
Note:
- type \((r,s)\) uses the same ordering as type \((q,r)\) with \(q = r\) and the second slot equal to \(s\).
- type \((1,0)\) is a vector.
- type \((0,1)\) is a covector.
- type \((0,1)\) is also called a \(1\)-form.
- the space of such tensors at a point \(P\) is denoted \(T^{r}_{\ s,P}(M)\).
definition [d] (Type-\((q,r)\) Tensor = Rank-\(n\) Tensor = Tensorial Set) A system of components with \(r\) upper and \(s\) lower indices that transform by the tensor law under coordinate changes,
- \(T^{\mu_1\ldots\mu_r}_{\ \nu_1\ldots\nu_s}\) ,
with total rank \(n = r + s\).
where
- \(T^{\mu_1\ldots\mu_r}_{\ \nu_1\ldots\nu_s}\) are the components of the tensor.
- \(\mu_1,\ldots,\mu_r\) are contravariant indices.
- \(\nu_1,\ldots,\nu_s\) are covariant indices.
- \(r\) is the contravariant rank.
- \(s\) is the covariant rank.
- \(n\) is the total rank.
Note:
- \(n\) is also called the order.
- upper indices transform contravariantly; lower indices covariantly.
- also called valence \(\{r,s\}\) in some texts.
196.1 Elementary Example
196.1.1 Simple
A type-\((0,1)\) tensor is a covector: one lower index and no upper index.
\[ T : V \rightarrow \mathbb{R} \]
\[ V = \{ e_{1},\ e_{2},\ e_{3} \} \]
\[ T(e_{1}) = 1,\quad T(e_{2}) = 0,\quad T(e_{3}) = -1 \]
\[ r = 0,\quad s = 1 \]
where
- \(r\) is the contravariant degree.
- \(s\) is the covariant degree.
196.1.2 General
A type-\((1,1)\) tensor on \(\mathbb{R}^{3}\) is a \(3 \times 3\) matrix acting as a linear map on components.
\[ T : V^{*} \times V \rightarrow \mathbb{R} \]
\[ (T^{i}_{\ j}) = \begin{pmatrix} 1 & 0 & 0 \\ 0 & 2 & 0 \\ 0 & 0 & 3 \end{pmatrix} \]
\[ T^{i}_{\ j}\, v^{j} = w^{i} \]
where
- \(T^{i}_{\ j}\) are the mixed components of a type-\((1,1)\) tensor.
- \(r = 1\) and \(s = 1\).
196.2 References
- Hassani, S. Mathematical Physics, 2nd ed. Springer. — type-\((r,s)\) tensor as multilinear map \(V^{*r}\times V^{s}\to\mathbb{R}\).
- Arfken, G. B., Weber, H. J., & Harris, F. E. Mathematical Methods for Physicists, 7th ed. Elsevier / Academic Press, 2013. — rank-\(n\) tensors; component transformation laws.
- Frankel, T. The Geometry of Physics, 3rd ed. Cambridge University Press. — multilinear tensor algebra on manifolds.
- Alternating Function
- Alternating k-linear function
- Boundary Operation
- Bundle
- Cartesian Product
- Constant Metric Tensor
- Contravariant Rank
- Covariant Rank
- Covariant, Contravariant, and Coordinate Systems
- Differential Forms
- Differential k-Form
- Edge of a Domain
- Exterior Algebra and Exterior Derivatives
- Forms
- Functions
- Inverse Metric
- k-Covector
- k-Covector Field
- k-Fold Product
- k-form
- k-Tensor
- Lie
- Lie Algebra
- Line Element
- Linear Functional
- Lorentzian Manifold
- Metric Tensor
- Multilinear Function
- Nondegenerate
- Nondegenerate Bilinear Form
- One-Form
- Smooth
- Smooth Assignment
- Smooth Mapping
- Stoke’s Theorem and the Fundamental Theorem of Calculus
- Symmetric Array
- Tangent Space
- Tensor Field
- Tensors
- Two-Form
- Type-(0,2) Tensor Field
- Type-(q,r) Tensor