15.20.4 Tensor Symmetric Tensor Operator Role
The symmetric tensor operator plays a key role in algebra by preserving symmetry and enabling structured transformations in tensor spaces.
Tensor Symmetric Tensor Operator Role is the identification of a symmetric tensor itself as a linear operator, mapping symmetric tensors of one degree to symmetric tensors of another degree via partial contraction, and it is this operator perspective, rather than the tensor's status as a static array of components, that underlies the catalecticant matrices and apolarity pairings central to symmetric tensor decomposition.
A Symmetric Tensor as a Map Between Graded Pieces
Partial Contraction as an Operator
Given a symmetric tensor T of order d on a vector space V, contracting T against k of its d slots using an element of the symmetric power S^k V-star, built from the dual space, produces a symmetric tensor of the remaining order d minus k. This assignment, sending an element of S^k V-star to the resulting element of S^{d-k} V, is linear, and it is exactly the operator, called the k-th catalecticant map of T, that is diagonalized or rank-analyzed throughout Reconstruction:
Every fact used about catalecticant matrices, such as their rank supplying a lower bound for symmetric rank as invoked under the Rank Relation, is a fact about this operator, not merely about a numerical array associated with T.
Recovering Familiar Constructions as Special Cases
Setting k equal to d recovers the full evaluation of T against d copies of a covector, an operator with one-dimensional domain-independent output matching the associated homogeneous polynomial discussed under the Polynomial Role; setting k equal to one recovers the gradient-type map used to define the quadratic form's associated linear map under the Symmetric Tensor Role, generalized here to arbitrary order.
The Symmetrization Operator
Symmetrization as a Projection
Beyond catalecticant maps built from a fixed tensor T, the symmetrization operator itself, sending an arbitrary tensor of order d to its symmetric part by averaging over the permutation action, is the primary example of a symmetric-tensor-related operator that does not depend on any particular T, but acts uniformly on the entire tensor power space; this operator is idempotent, since applying it twice produces the same result as applying it once, exactly the defining property of a projection, and its image is precisely the symmetric subspace whose invariance is established under Subspace Invariance.
Bracket Notation as Operator Notation
The Symmetrization Bracket Notation introduced earlier is, in this Operator Role perspective, notation for the output of applying the symmetrization operator to a specified subset of indices, and the partial symmetrization variants, using vertical bars to exclude certain indices, correspond to applying a projection operator that acts nontrivially only on a chosen subset of tensor factors while leaving the remainder fixed.
Apolarity as an Operator Pairing
Differential Operators Dual to Vectors
Through the Polynomial Role's identification of symmetric tensors with homogeneous polynomials, a covector w in the dual space of V corresponds to a constant-coefficient differential operator, and applying this differential operator to the homogeneous polynomial associated with T a total of k times realizes exactly the k-th catalecticant operator applied to w raised to the symmetric power k; the apolar ideal of T, central to Reconstruction, is defined as the set of covectors (and, more generally, higher-degree elements of the dual symmetric algebra) whose associated differential operator annihilates the polynomial of T, making apolarity itself a statement about the kernel of this family of operators.
Operator Rank and Decomposition
The rank of the k-th catalecticant operator, for the value of k nearest to d divided by two, generally supplies the sharpest lower bound on the symmetric rank of T obtainable by linear-algebraic means, since a genuine decomposition of T into r pure power forms forces every catalecticant operator associated with T to factor through an r-dimensional intermediate space, bounding its rank by r; this operator-theoretic argument is the precise mechanism, described only informally elsewhere, by which catalecticant rank bounds are proven.
Consequences of the Operator Perspective
Unifying Apolarity, Catalecticants, and Symmetrization
Recognizing that catalecticant maps, the symmetrization projector, and apolar differential operators are all instances of the same general idea, a symmetric tensor or a fixed differential structure acting as a linear operator on a graded piece of the symmetric algebra, unifies constructions that might otherwise appear as an unrelated list of ad hoc tools within Reconstruction and rank theory, situating them instead as concrete manifestations of the single Operator Role a symmetric tensor plays within the broader symmetric algebra.
Complementing the Representation, Polynomial, Form, and Geometry Roles
Where the Representation Role treats S^d V as an object acted upon by an external group, the Operator Role treats a chosen element of a symmetric power itself as the acting agent, mapping between graded pieces; together with the Polynomial, Form, and Geometry Roles, the Operator Role completes the account of how symmetric tensors function algebraically well beyond their basic definition via the Component Constraint.