1.2.63 Tensor Higher Order Component Definition

In tensor algebra, higher order components define structured data relationships through multi-linear mappings, essential for advanced mathematical modeling.

Tensor Higher Order Component Definition is the specification of the individual numerical entries that make up the component array of a tensor whose rank exceeds two, meaning the total number of upper and lower indices, p + q, is three or greater, generalizing the familiar notions of vector components, covector components, and matrix-like components to arrays that can no longer be displayed as a simple list or a flat grid. A tensor higher order component is addressed by three or more independent index values simultaneously, each ranging over the dimension of the underlying vector space, and it represents one entry within a structure that must be understood combinatorially rather than visually.

Definition and Index Structure

General Form

A tensor T of type (p, q) with p + q equal to three or more has components written with p upper indices and q lower indices, all appearing simultaneously on a single symbol.

T_{j_{1} \dots j_{q}}^{i^{1} \dots i^{p}}

Each of the p + q indices independently ranges from 1 to n, where n is the dimension of the vector space, and a higher order component is obtained by fixing a specific value for every one of these indices at once.

Total Component Count

The total number of higher order components equals n raised to the power of the rank, matching the general formula used for tensors of any rank, but growing rapidly once the rank exceeds two.

number of components = n^{p + q}

For instance, a rank-three tensor in a four-dimensional space has 4^3 = 64 components, and a rank-four tensor in the same space has 4^4 = 256 components.

Rank Three Components

Structure as a Stack of Matrices

A rank-three tensor component, such as T^i_{jk} for a type (1, 2) tensor, can be organized conceptually as a stack of two-dimensional matrices, one matrix for each value of the free index that is not part of the pair forming the matrix. Fixing the third index selects one matrix from the stack, and the remaining two indices locate an entry within that matrix exactly as in the rank-two case.

Example: The Structure Constants

A common example of a rank-three tensor component arises in the structure constants of a Lie algebra, written f^i_{jk}, which encode how the bracket of two basis elements decomposes back into the basis, with one upper index for the resulting basis direction and two lower indices for the pair of basis elements being combined.

Rank Four and Beyond

Structure as a Grid of Stacks

A rank-four tensor component can be organized as a two-dimensional grid whose entries are themselves stacks of matrices, effectively nesting the rank-three structure one level deeper. Continuing this pattern, a rank-k tensor component array can always be understood as nested collections of lower-rank arrays, one level of nesting for each additional index beyond two.

Example: The Riemann Curvature Tensor

A prominent example of a rank-four tensor component is the Riemann curvature tensor, typically written R^i_{jkl}, which measures the failure of parallel transport to return a vector to its original value after traversing a closed loop, and which requires four independent indices to encode the two directions defining the loop together with the two directions describing the vector being transported.

Symmetry Patterns in Higher Order Components

Symmetry Among Subsets of Indices

Higher order tensors frequently exhibit symmetry or antisymmetry not across all of their indices at once but across specific subsets. A component may be symmetric under the exchange of two particular indices while showing no special relationship to a third.

T_{i j k} = T_{j i k}

indicates symmetry in the first two indices only, leaving the relationship between either of those indices and the third index unconstrained.

Total Antisymmetry

A higher order component may instead be totally antisymmetric across all of its indices, changing sign under the exchange of any pair, which is the defining property of the components of a differential form of degree equal to the number of indices involved.

Diagrammatic Summary

The diagram represents a rank-three component array as three overlapping matrices, each one corresponding to a fixed value of the third index, illustrating how a higher order component is located by first selecting a matrix in the stack and then locating an entry within it.