Integral projective tensor product is a Banach space concept that enriches the classical projective tensor product with integral representations and measure-theoretic techniques.
The approach integrates bilinear forms, integral operators, and Bochner integrals to streamline norm attainment and duality analysis in tensor products.
It finds application in operator algebras, C*-algebras, and protoquantum spaces, influencing convex geometry, asymptotic structure, and ideal theory.
The integral projective tensor product is not presented in the surveyed literature as a single universally fixed tensor norm. The stable background object is the completed projective tensor product X⊗πY of Banach spaces, defined by
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},
while “integral” enters through several closely related mechanisms: integral bilinear forms and integral operators, Bochner-integral representations of projective norm-attaining tensors, and duality with injective-type constructions (Aliaga et al., 27 Feb 2026, Kumar et al., 2013).
1. Terminology and scope
In Banach-space theory, the phrase belongs to the orbit of projective tensor products, integral operators, and measure-theoretic representations. It does not refer to the algebro-geometric usage of “tensor product surface” in which a surface is parametrized from P1×P1 by bihomogeneous forms; that terminology is explicitly unrelated to the integral/projective tensor product of Banach space theory (Duarte et al., 2019).
Within the C∗-algebraic and operator-space literature surveyed here, an important terminological point is that several papers study the Banach-space projective tensor product A⊗γB or the operator-space projective tensor product A⊗B, but do not introduce a separate tensor norm called the integral tensor product. Instead, “integral” appears through integral bilinear forms and integral operators, which are used to describe canonical dual embeddings and comparison maps (Kumar et al., 2013, Gupta et al., 2018).
A second terminological distinction concerns recent work on norm attainment. There, “integral projective norm-attaining tensor” denotes a tensor in X⊗πY represented by a Bochner integral over BX×BY with total mass equal to the projective norm; this is a measure-theoretic enlargement of the classical class of norm-attaining tensors, not a replacement for the underlying projective tensor norm itself (Aliaga et al., 27 Feb 2026).
2. Classical projective tensor products and their duality
For Banach spaces X and Y, the projective tensor product ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},0 is the completion of the algebraic tensor product under ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},1. Its dual admits the standard identification
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},2
and the pairing is given on elementary tensors by ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},3 (Aliaga et al., 27 Feb 2026). This duality is the basic reason projective tensor products linearize bounded bilinear forms.
The same pattern persists in higher tensor powers. For ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},4, the ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},5-fold projective tensor product ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},6 is defined recursively, and its dual is the Banach space of bounded ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},7-linear forms: ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},8
In the case ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},9, the projective tensor tower is especially rigid: for P1×P10, P1×P11 is not isomorphic to any subspace of any quotient of P1×P12, and
An analogous projective formalism exists beyond Banach spaces. For Banach P1×P14-modules P1×P15 and P1×P16, the projective pointwise norm is
P1×P17
the completed projective tensor product is P1×P18, and its universal property is
P1×P19
Its dual is correspondingly identified as
C∗0
(Pasqualetto, 2023). This C∗1-valued construction preserves the classical projective theme while changing both the scalar ring and the norm range.
3. Integral operators, bilinear forms, and projective tensor products
A central bridge between “integral” and “projective” is duality. For C∗2-algebras C∗3 and C∗4, the canonical map
C∗5
is bounded above and below by
C∗6
and C∗7 is identified in the paper with a space of integral operators (Kumar et al., 2013). The same paper defines a natural embedding
C∗8
with the same bicontinuity estimate,
C∗9
Under additional hypotheses it records the coincidence
A⊗γB0
linking nuclear, Pietsch integral, and integral operators.
This operator-theoretic perspective complements the usual projective-tensor description by bilinear forms. For Banach algebras A⊗γB1 and A⊗γB2, Arens regularity of A⊗γB3 is equivalent to the statement that every bounded bilinear form A⊗γB4 is biregular, while Arens regularity of the operator-space projective tensor product A⊗γB5 is equivalent to biregularity of every jointly completely bounded bilinear form (Kumar et al., 2014). In other words, projective tensor products are classified by the classes of bilinear forms they linearize, and integral-operator language enters through the dual embeddings that control these spaces.
The same dual mechanism appears in recent work on norm attainment. If a functional on A⊗γB6 attains its norm at an integral norm-attaining tensor, then the corresponding bilinear form is norm-attaining on A⊗γB7 (Aliaga et al., 27 Feb 2026). This relation is one reason integral representations are useful: they convert extremal questions in the tensor product into norm-attainment questions for bilinear forms and operators.
4. Integral projective norm-attaining tensors
The most explicit recent integral formulation is the class A⊗γB8 of integral projective norm-attaining tensors. A tensor A⊗γB9 belongs to this class if there exists a finite positive Borel measure A⊗B0 on A⊗B1 such that the canonical map
A⊗B2
is A⊗B3-Bochner integrable and
A⊗B4
Classical norm-attaining tensors yield such representations by taking a discrete measure, so
A major structural theorem states that every integral norm-attaining tensor can be approximated in projective norm by finitely norm-attaining tensors: X⊗πY4
Consequently, the Bishop–Phelps density problem is unchanged by passing from discrete to integral representations: X⊗πY5
The same paper proves that if an extreme point of X⊗πY6 belongs to X⊗πY7 and is witnessed by a Radon measure, then it must be an elementary tensor X⊗πY8 (Aliaga et al., 27 Feb 2026).
Weak and weakX⊗πY9 variants are developed by replacing the norm-topology Borel structure on BX×BY0 with weak or weakBX×BY1 product topologies. Most formal properties persist, but genuine Bochner integrability becomes delicate. Under separability together with either BAP or a Dunford–Pettis-type hypothesis, the tensor-valued map BX×BY2 is Bochner integrable in the weak setting. The same work also extends known non-norm-attainment phenomena to the integral setting, showing for instance that BX×BY3 and the real BX×BY4 contain non-norm-attaining tensors for BX×BY5 (Aliaga et al., 27 Feb 2026).
5. Geometric and asymptotic structure
Projective tensor products have a rigid convex geometry. For bounded closed convex sets BX×BY6 and BX×BY7, the set
BX×BY8
is the natural tensorial convex hull, and when BX×BY9 separates points of X0, every nonzero preserved extreme point of this set is of the form X1, with X2 and X3 preserved extreme in X4 and X5 respectively (García-Lirola et al., 2022). Applied to unit balls, this yields a factorization theorem for preserved extreme points of X6.
The same paper gives a partial converse for weak-strongly exposed points. Under the additional assumption that X7 has a compact neighborhood system for the weak topology in X8, weak-strong exposure tensorizes: X9
The extra hypothesis is essential. There exists a Banach space Y0 isomorphic to Y1 with a weak-strongly exposed point Y2 such that Y3 is not weak-strongly exposed in Y4 (García-Lirola et al., 2022).
Asymptotically, iterated projective tensor products form a strict hierarchy. For Y5,
Y6
is not isomorphic to any subspace of any quotient of
Y7
This is stronger than mere non-isomorphism and is proved through asymptotic invariants such as the classes Y8 and Szlenk-type bounds (Causey et al., 2020). A plausible implication is that projective tensoring increases asymptotic complexity in a systematic way rather than merely enlarging dimension or codimension.
These geometric results interact naturally with the integral perspective. The recent extremal theorem for Radon-integral norm-attaining tensors gives a measure-theoretic route to the same rank-one conclusion that older convex-geometric arguments derive from compact-operator separation (Aliaga et al., 27 Feb 2026, García-Lirola et al., 2022). This suggests a convergence of two lines of thought: one through weak topology and operator ideals, the other through Bochner representations.
6. Y9-algebraic and operator-space variants
For ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},00-algebras, the Banach-space projective tensor product is denoted ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},01. It is a Banach ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},02-algebra, and its ideal theory is unusually explicit. If ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},03 and ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},04 are ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},05-subalgebras, the identity map on ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},06 extends to an isometric ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},07-algebra map from ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},08 onto the closed ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},09-subalgebra ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},10 (Gupta et al., 2018). When one factor is topologically simple, every closed ideal of ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},11 is of the form ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},12. More generally, if one factor has finitely many closed ideals, every closed ideal is a finite sum of product ideals. The same paper identifies maximal ideals by
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},13
for maximal ideals ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},14, ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},15, and proves the center formula
The operator-space projective tensor product uses a different norm. For operator spaces ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},17 and ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},18,
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},19
with ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},20, ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},21, ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},22, and ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},23 (Kumar et al., 2014). In the ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},24-algebraic setting, the canonical embedding
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},25
is ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},26-preserving, positive, completely bounded, and satisfies
Arens regularity sharply links the Banach and operator-space projective theories. For ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},28-algebras ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},29 and ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},30, the Arens regularity of
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},31
is equivalent (Kumar et al., 2014). Thus, at the level of bidual multiplication, classical projective, operator-space projective, Haagerup, and Schur tensor norms exhibit the same regularity behavior in the ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},32-algebraic category.
The Banach ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},33-module setting extends projective tensor products to norms with values in the ring ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},34. There the projective tensor product ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},35 is defined by completion of the algebraic tensor product over ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},36, and its dual satisfies
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},37
It also obeys a series representation formula
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},38
and a pullback compatibility theorem
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},39
for measurable maps ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},40 under the stated hypotheses (Pasqualetto, 2023). This is a projective tensor theory over a function ring rather than over ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},41 or ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},42.
An even more noncommutative extension is the projective tensor product of protoquantum spaces. For PQ-spaces ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},43 and ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},44, the proto-operator-projective norm on ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},45 is
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},46
where the infimum runs over decompositions
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},47
The resulting completed tensor product ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},48 linearizes completely bounded bilinear maps and satisfies an adjoint associativity theorem
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},49
(Helemskii, 2017). In this framework, the standard operator-space projective formula is too rigid outside the operator-space category: the paper exhibits tensors ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},50 for which
∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},51
This shows that a projective tensor product suitable for general matricially normed spaces cannot simply be the usual operator-space projective tensor product (Helemskii, 2017).
Taken together, these developments show that “integral projective tensor product” names not a single immutable object but a family of tightly connected ideas. The fixed core is the projective tensor product and its universal property; the integral aspect appears through integral operators, bilinear forms, and, most explicitly, through Bochner-integral representations of norm-attaining tensors. The interaction between these themes governs extremal geometry, non-attainment phenomena, Arens regularity, ideal structure, and the extension of projective tensor methods to operator spaces, ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},52-algebras, ∥u∥π=inf{i=1∑n∥xi∥∥yi∥:u=i=1∑nxi⊗yi},53-modules, and protoquantum spaces.