7.13.2 Tensor Component Index Selection

Tensor Component Index Selection is the step within the extraction procedure at which specific numerical values are assigned to each of a tensor's index positions, determining precisely which basis elements will be supplied as arguments and, therefore, which single entry of the component table is being computed.

The Role of Index Selection in Extraction

Choosing a Point Within the Index Range

Before extraction can proceed, a definite value must be selected for every index the tensor carries, with each selected value drawn from the range of one to the dimension of the underlying space, together fixing exactly one index tuple out of all those available.

Translating Selected Values Into Arguments

Once numerical values are selected for each index, those values determine exactly which basis vector or dual basis covector must be supplied to the corresponding slot of the tensor, converting an abstract selection of numbers into a concrete evaluation instruction.

Selection Strategies

Exhaustive Selection for a Complete Table

To build the entire component table, index selection is repeated for every possible combination of index values, visiting each index tuple in turn, typically following a systematic order such as lexicographic order, to guarantee that no entry is missed and none is computed twice.

Targeted Selection for a Single Entry

When only one particular entry is of interest, such as checking a suspected symmetry relation or verifying a specific numerical prediction, index selection can instead be applied just once, to the single tuple relevant to the question at hand, avoiding the unnecessary work of computing the full table.

Constraints on Valid Selection

Respecting the Dimension of the Space

Every selected index value must fall within the valid range determined by the dimension of the underlying vector space; selecting a value outside this range corresponds to no basis element at all and renders the intended extraction undefined.

Respecting the Tensor's Declared Type

The number of index positions requiring a selected value is fixed by the tensor's rank, and the variance assigned to each position is fixed by the tensor's declared type, so index selection must supply exactly the right number of values, each destined to choose an ordinary basis vector or a dual basis covector according to its position's variance.

Selection and Repeated Indices

Selecting the Same Value Across Multiple Positions

Nothing in the general procedure forbids selecting the same numerical value for two or more different index positions, and doing so is precisely what is needed, for example, to extract a diagonal entry of a matrix component case or to test whether an antisymmetric tensor correctly produces a vanishing value at such a tuple.

Selection Underlying Summation Conventions

When the Einstein summation convention is used, a repeated index symbol appearing once as an upper position and once as a lower position signals that index selection is to be performed for every valid value of that shared symbol in turn, with the resulting terms summed together rather than any single value being fixed.

Diagrammatic Illustration

Selecting specific index values narrows the full space of possible tuples down to the single tuple that will actually be evaluated.

Significance for the Broader Extraction Process

Precondition for Meaningful Evaluation

Index selection must occur before the actual evaluation step of extraction can proceed, since the tensor cannot be applied to unspecified or partially specified arguments; selection is what turns the general extraction operation into a fully determined instruction ready to be carried out.

Interaction With the Component Table's Structure

The pattern of index selections carried out across an entire extraction campaign directly determines the shape and completeness of the resulting component table, making disciplined, systematic index selection essential whenever a full and accurate table representation of the tensor is required.