7.3 Tensor Component Structure

Tensor Component Structure is the overall arrangement by which the individual numerical entries of a tensor are organized, indexed, and related to one another once a basis has been chosen. It covers how many components a tensor has, how those components are laid out across upper and lower indices, and what patterns of dependence, symmetry, or redundancy connect them.

Definition and Scope

Index Arrangement

A tensor of type ((p,q)) on an (n)-dimensional vector space has components carrying (p) upper indices and (q) lower indices, each running from (1) to (n). The component structure specifies this arrangement directly:

giving a total of (n^{p+q}) entries before any symmetry is taken into account.

Layout as a Multidimensional Array

Concretely, the component structure of a rank-(k) tensor is a (k)-dimensional array, with each dimension of size (n). A rank-2 tensor is laid out as a matrix, a rank-3 tensor as a cube of numbers, and higher-rank tensors as the natural generalization, each axis of the array corresponding to one index of the tensor.

Structural Properties

Independent Versus Dependent Components

Not every entry in the array is independent. Symmetries, antisymmetries, and trace conditions reduce the number of independent components below the raw count of (n^{p+q}). A symmetric rank-2 tensor, for example, has

independent components rather than (n^2), and the component structure records both the full array and the reduced description implied by such constraints.

Blocks and Sub-Structures

Component structure can be organized hierarchically. A rank-4 tensor may be viewed as a square array of rank-2 blocks, each block itself a matrix, with the component structure describing both the outer arrangement of blocks and the inner arrangement within each block. This layered view is common when a tensor arises from combining several lower-rank tensors, such as a tensor product.

Basis Dependence of the Structure

The component structure exists only relative to a chosen basis; changing basis changes every entry of the array according to the tensor's transformation law, while the shape and index arrangement, the number of upper and lower indices and the dimension (n), remain fixed regardless of basis.

Role Within Tensor Algebra

Foundation for Operations

Every algebraic operation on tensors, addition, scalar multiplication, tensor product, and contraction, is defined directly in terms of component structure: addition combines entries at matching index positions, contraction sums over matched upper and lower indices, and the tensor product concatenates the index structures of its factors.

Storage and Computation

In computational settings, component structure determines memory layout: contiguous versus strided storage, sparse versus dense representation, and whether symmetry can be exploited to store only independent components. Understanding the component structure of a tensor in advance allows algorithms to choose storage and computation strategies that avoid redundant work.