6.2.2 Tensor Degree Classification Area

Tensor Degree Classification Area explores how tensors are categorized by their degree, defining their structure and operations in algebraic frameworks.

Tensor Degree Classification Area is the branch of tensor classification that locates a tensor within the graded structure of the full tensor algebra T(V) = ⊕_{n≥0} V^{⊗n}, assigning to each homogeneous element the integer n marking which direct summand it belongs to, and extending this assignment to general, non-homogeneous elements by decomposing them into a finite sum of homogeneous parts of different degrees. Where order classification describes a single tensor's index count viewed in isolation, degree classification describes the same integer viewed as a grading label within the ambient algebra that contains every order of tensor built from V at once.

The Graded Structure of the Tensor Algebra

Direct Sum Decomposition by Degree

The tensor algebra built from a vector space V decomposes as a direct sum indexed by degree:

T (V) = \oplus_{n = 0}^{\infty} V^{\otimes n}

with V^{⊗0} = F, V^{⊗1} = V, V^{⊗2} = V ⊗ V, and so forth; every element of T(V) lies, or decomposes uniquely into parts that each lie, in exactly one of these summands.

Homogeneous Elements

An element of T(V) is homogeneous of degree n if it lies entirely within the summand V^{⊗n}; every simple tensor v₁ ⊗ v₂ ⊗ ... ⊗ v_n is homogeneous of degree n by construction, since the tensor product of n vectors from V produces an element of V^{⊗n} and nothing else.

General Elements as Finite Sums of Homogeneous Parts

An arbitrary element of T(V) need not be homogeneous; it is instead a finite sum x = x₀ + x₁ + x₂ + ... + x_m with each x_i ∈ V^{⊗i}, and degree classification assigns such an element the full list of degrees appearing with a nonzero term, rather than a single number.

Diagram of the Graded Decomposition

Degree Arithmetic Under Algebra Multiplication

Additivity of Degree Under the Algebra Product

The multiplication of the tensor algebra, given by concatenating tensor factors, sends a degree-m element and a degree-n element to a degree-(m+n) element:

x \in V^{\otimes m}, y \in V^{\otimes n} \Rightarrow x y \in V^{\otimes (m + n)}

which is exactly what makes T(V) a graded algebra rather than merely a direct sum of unrelated vector spaces.

Degree of a Sum Within the Same Summand

Two homogeneous elements of the same degree n add to another homogeneous element of degree n, since V^{⊗n} is a vector space and closed under its own addition; degree is preserved by linear combination only when every term shares the same degree.

Loss of Homogeneity When Degrees Differ

Adding two homogeneous elements of different degrees, such as a degree-1 vector and a degree-2 simple tensor, produces an element that is not homogeneous at all, but is instead classified by degree as having a degree-1 part and a degree-2 part simultaneously.

Distinguishing Degree From Order

Order Describes One Tensor; Degree Locates It in an Algebra

Order classification is a property computed directly from a single tensor's own indices, independent of any surrounding algebraic structure; degree classification is the same number reinterpreted as a coordinate identifying which summand of the ambient graded algebra T(V) a homogeneous tensor occupies.

Degree Applies Naturally to Sums; Order Does Not

Because order is defined per individual tensor, the phrase "the order of x₁ + x₂" is only meaningful when x₁ and x₂ share the same order; degree classification, by contrast, is designed from the outset to describe general, inhomogeneous sums as elements with parts of several different degrees, since the tensor algebra is built as a direct sum precisely to accommodate this.

Both Coincide on Homogeneous Simple Tensors

For a homogeneous simple tensor v₁ ⊗ ... ⊗ v_n, the order (as an index count) and the degree (as a grading label) are numerically the same value n; the two classification areas diverge only once general, non-simple, or inhomogeneous elements of the tensor algebra are considered.

Why Degree Classification Is Useful

Organizing Algebraic Manipulations by Homogeneous Parts

Many identities and computations within T(V) are proved by reducing to the homogeneous case and then extending linearly, a strategy that depends on the ability to isolate an element's degree-n part, which degree classification makes precise.

Foundation for Grading-Based Constructions

Constructions such as the symmetric algebra and the exterior algebra are obtained from T(V) by taking quotients or subspaces degree by degree, so that the resulting algebra inherits its own grading directly from the degree classification already fixed on T(V).