15.14 Tensor Symmetric Decomposition Structure

Tensor Symmetric Decomposition Structure is the overarching framework concerned with expressing a totally symmetric tensor as a combination of simpler, more elementary symmetric tensors, most centrally as a sum of pure power terms, and with the collection of concepts, the term itself, the minimality condition, the decomposition count, and the governing decomposition relation, that together make this expression precise, well-defined, and computable. This structure sits above the symmetric rank structure as its organizing context, situating the rank invariant within the broader set of questions any decomposition-based study of symmetric tensors must address: what building blocks are permitted, what equation must a decomposition satisfy, how few building blocks suffice, and how unique is the resulting expression.

Framing decomposition as its own structure, distinct from but built upon the rank structure specifically, acknowledges that decomposition raises questions beyond the single numerical invariant of rank, including the explicit equation a decomposition must solve and the broader question of when and how uniquely such decompositions exist, treating rank as one central output of a larger decomposition-theoretic inquiry rather than as the entirety of that inquiry.

The Decomposition Question in General

What Counts as a Valid Decomposition

A decomposition of a symmetric tensor T, in the sense developed throughout this structure, is any finite expression of T as a sum of pure power terms, scalar multiples of symmetric powers of individual vectors, satisfying the decomposition relation exactly at every independent component of T; this defines the class of admissible decompositions before any question of minimality or efficiency is imposed.

Why Pure Powers Are the Chosen Building Blocks

Pure power terms are singled out as the elementary units of decomposition because they correspond, under the polynomial analogy, to the simplest possible nonzero homogeneous polynomials, perfect powers of a single linear form, making them the natural atomic pieces from which more complex homogeneous polynomials, and hence more complex symmetric tensors, can be assembled.

The Governing Equation and Its Solvability

The Decomposition Relation as the Central Equation

Every discussion of decomposition ultimately reduces to solving the decomposition relation, the explicit system of equations, linear in the scalar coefficients but nonlinear in the chosen vectors, that a candidate collection of pure power terms must satisfy to reconstruct a given tensor T exactly.

Rank as the Threshold of Solvability

The symmetric rank of T is precisely the smallest number of terms for which this system first becomes solvable, connecting the abstract decomposition relation directly to the concrete, well-defined invariant established through the minimality condition and quantified through the decomposition count.

From Existence to Minimality to Uniqueness

A Layered Set of Questions

The decomposition structure organizes its central questions in increasing order of difficulty: first, whether a decomposition exists at all for a given number of terms, a question always answerable affirmatively once enough terms are allowed, by the spanning property of pure power terms; second, what the minimal such number of terms is, addressed by the rank structure; and third, whether a minimal decomposition, once found, is unique up to the unavoidable scalar and sign ambiguities inherent to any single pure power term.

Uniqueness as the Frontier of the Structure

Uniqueness of minimal decompositions, established generically for many combinations of rank and dimension but failing on special exceptional loci, represents the most delicate and least universally resolved question within the decomposition structure, marking the boundary between fully understood cases, such as rank-two tensors reducible to ordinary matrix decomposition, and genuinely open or actively studied cases at higher rank.

Placement Within the Broader Symmetric Tensor Framework

Complementary to Representation-Oriented Structures

Where the independent symmetric component structure and the symmetric basis structure address how to represent a symmetric tensor faithfully and efficiently with respect to a fixed reference, the symmetric decomposition structure addresses how to represent the same tensor using the fewest possible elementary, basis-independent pieces, offering a fundamentally different, complexity-and-efficiency-oriented perspective on the same underlying mathematical objects.

A Natural Endpoint of the Developed Theory

Because decomposition draws on nearly every earlier construction, the symmetric power as the elementary unit, the polynomial analogy for translating between tensor and polynomial language, and the count relation for bounding the scale of the problem, the symmetric decomposition structure functions as a natural culminating topic, synthesizing the algebraic machinery built up across the theory of symmetric tensors into the single, unifying question of how economically a symmetric tensor can be expressed.