Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Tensor product
open-in-new
<h2 id="definitions-and-constructions">Definitions and constructions</h2> <p>The <i>tensor product</i> of two vector spaces is a vector space that is defined <a href="/facts/Up_to/ydz4ztzX">up to</a> an <a href="/facts/Isomorphism/pj8KgkxU">isomorphism</a>. There are several equivalent ways to define it. Most consist of defining explicitly a vector space that is called a tensor product, and, generally, the equivalence proof results almost immediately from the basic properties of the vector spaces that are so defined. </p><p>The tensor product can also be defined through a <a href="/facts/Universal_property/5XJBYMNz">universal property</a>; see § Universal property, below. As for every universal property, all <a href="/facts/Object_(category_theory)/7WzxUThf">objects</a> that satisfy the property are isomorphic through a unique isomorphism that is compatible with the universal property. When this definition is used, the other definitions may be viewed as constructions of objects satisfying the universal property and as proofs that there are objects satisfying the universal property, that is that tensor products exist. </p> <h3>From bases</h3> <p>Let V and W be two <a href="/facts/Vector_space/M1pxMLx2">vector spaces</a> over a <a href="/facts/Field_(mathematics)/xAjAS4ko">field</a> F, with respective <a href="/facts/Basis_(linear_algebra)/89IPoN6c">bases</a> B V {\displaystyle B_{V}} and B W {\displaystyle B_{W}} . </p><p>The <i>tensor product</i> V ⊗ W {\displaystyle V\otimes W} of V and W is a vector space that has as a basis the set of all v ⊗ w {\displaystyle v\otimes w} with v ∈ B V {\displaystyle v\in B_{V}} and w ∈ B W {\displaystyle w\in B_{W}} . This definition can be formalized in the following way (this formalization is rarely used in practice, as the preceding informal definition is generally sufficient): V ⊗ W {\displaystyle V\otimes W} is the set of the <a href="/facts/Function_(mathematics)/CdReEKCh">functions</a> from the <a href="/facts/Cartesian_product/BfqBWzdL">Cartesian product</a> B V × B W {\displaystyle B_{V}\times B_{W}} to F that have a finite number of nonzero values. The <a href="/facts/Pointwise_operation/dt4Mrs6w">pointwise operations</a> make V ⊗ W {\displaystyle V\otimes W} a vector space. The function that maps ( v , w ) {\displaystyle (v,w)} to 1 and the other elements of B V × B W {\displaystyle B_{V}\times B_{W}} to 0 is denoted v ⊗ w {\displaystyle v\otimes w} . </p><p>The set { v ⊗ w ∣ v ∈ B V , w ∈ B W } {\displaystyle \{v\otimes w\mid v\in B_{V},w\in B_{W}\}} is then straightforwardly a basis of V ⊗ W {\displaystyle V\otimes W} , which is called the <i>tensor product</i> of the bases B V {\displaystyle B_{V}} and B W {\displaystyle B_{W}} . </p><p>We can equivalently define V ⊗ W {\displaystyle V\otimes W} to be the set of <a href="/facts/Bilinear_form/gyvzn9ym">bilinear forms</a> on V × W {\displaystyle V\times W} that are nonzero at only a finite number of elements of B V × B W {\displaystyle B_{V}\times B_{W}} . To see this, given ( x , y ) ∈ V × W {\displaystyle (x,y)\in V\times W} and a bilinear form B : V × W → F {\displaystyle B:V\times W\to F} , we can decompose x {\displaystyle x} and y {\displaystyle y} in the bases B V {\displaystyle B_{V}} and B W {\displaystyle B_{W}} as: x = ∑ v ∈ B V x v v and y = ∑ w ∈ B W y w w , {\displaystyle x=\sum _{v\in B_{V}}x_{v}\,v\quad {\text{and}}\quad y=\sum _{w\in B_{W}}y_{w}\,w,} where only a finite number of x v {\displaystyle x_{v}} 's and y w {\displaystyle y_{w}} 's are nonzero, and find by the bilinearity of B {\displaystyle B} that: B ( x , y ) = ∑ v ∈ B V ∑ w ∈ B W x v y w B ( v , w ) {\displaystyle B(x,y)=\sum _{v\in B_{V}}\sum _{w\in B_{W}}x_{v}y_{w}\,B(v,w)} </p><p>Hence, we see that the value of B {\displaystyle B} for any ( x , y ) ∈ V × W {\displaystyle (x,y)\in V\times W} is uniquely and totally determined by the values that it takes on B V × B W {\displaystyle B_{V}\times B_{W}} . This lets us extend the maps v ⊗ w {\displaystyle v\otimes w} defined on B V × B W {\displaystyle B_{V}\times B_{W}} as before into bilinear maps v ⊗ w : V × W → F {\displaystyle v\otimes w:V\times W\to F} , by letting: ( v ⊗ w ) ( x , y ) := ∑ v ′ ∈ B V ∑ w ′ ∈ B W x v ′ y w ′ ( v ⊗ w ) ( v ′ , w ′ ) = x v y w . {\displaystyle (v\otimes w)(x,y):=\sum _{v'\in B_{V}}\sum _{w'\in B_{W}}x_{v'}y_{w'}\,(v\otimes w)(v',w')=x_{v}\,y_{w}.} </p><p>Then we can express any bilinear form B {\displaystyle B} as a (potentially infinite) formal linear combination of the v ⊗ w {\displaystyle v\otimes w} maps according to: B = ∑ v ∈ B V ∑ w ∈ B W B ( v , w ) ( v ⊗ w ) {\displaystyle B=\sum _{v\in B_{V}}\sum _{w\in B_{W}}B(v,w)(v\otimes w)} making these maps similar to a <a href="/facts/Schauder_basis/na8qTLT6">Schauder basis</a> for the vector space Hom ( V , W ; F ) {\displaystyle {\text{Hom}}(V,W;F)} of all bilinear forms on V × W {\displaystyle V\times W} . To instead have it be a proper Hamel <a href="/facts/Basis_(linear_algebra)/89IPoN6c">basis</a>, it only remains to add the requirement that B {\displaystyle B} is nonzero at an only a finite number of elements of B V × B W {\displaystyle B_{V}\times B_{W}} , and consider the subspace of such maps instead. </p><p>In either construction, the <i>tensor product of two vectors</i> is defined from their decomposition on the bases. More precisely, taking the basis decompositions of x ∈ V {\displaystyle x\in V} and y ∈ W {\displaystyle y\in W} as before: x ⊗ y = ( ∑ v ∈ B V x v v ) ⊗ ( ∑ w ∈ B W y w w ) = ∑ v ∈ B V ∑ w ∈ B W x v y w v ⊗ w . {\displaystyle {\begin{aligned}x\otimes y&={\biggl (}\sum _{v\in B_{V}}x_{v}\,v{\biggr )}\otimes {\biggl (}\sum _{w\in B_{W}}y_{w}\,w{\biggr )}\\[5mu]&=\sum _{v\in B_{V}}\sum _{w\in B_{W}}x_{v}y_{w}\,v\otimes w.\end{aligned}}} </p><p>This definition is quite clearly derived from the coefficients of B ( v , w ) {\displaystyle B(v,w)} in the expansion by bilinearity of B ( x , y ) {\displaystyle B(x,y)} using the bases B V {\displaystyle B_{V}} and B W {\displaystyle B_{W}} , as done above. It is then straightforward to verify that with this definition, the map ⊗ : ( x , y ) ↦ x ⊗ y {\displaystyle {\otimes }:(x,y)\mapsto x\otimes y} is a bilinear map from V × W {\displaystyle V\times W} to V ⊗ W {\displaystyle V\otimes W} satisfying the <a href="/facts/Universal_property/5XJBYMNz">universal property</a> that any construction of the tensor product satisfies (see below). </p><p>If arranged into a rectangular array, the <a href="/facts/Coordinate_vector/TEcLqS40">coordinate vector</a> of x ⊗ y {\displaystyle x\otimes y} is the <a href="/facts/Outer_product/qR9C2BU1">outer product</a> of the coordinate vectors of x {\displaystyle x} and y {\displaystyle y} . Therefore, the tensor product is a generalization of the outer product, that is, an abstraction of it beyond coordinate vectors. </p><p>A limitation of this definition of the tensor product is that, if one changes bases, a different tensor product is defined. However, the decomposition on one basis of the elements of the other basis defines a <a href="/facts/Canonical_isomorphism/pj8KgkxU">canonical isomorphism</a> between the two tensor products of vector spaces, which allows identifying them. Also, contrarily to the two following alternative definitions, this definition cannot be extended into a definition of the <a href="/facts/Tensor_product_of_modules/K3oaFIS4">tensor product of modules</a> over a <a href="/facts/Ring_(mathematics)/JnCUXz63">ring</a>. </p> <h3>As a quotient space</h3> <p>A construction of the tensor product that is basis independent can be obtained in the following way. </p><p>Let V and W be two <a href="/facts/Vector_space/M1pxMLx2">vector spaces</a> over a <a href="/facts/Field_(mathematics)/xAjAS4ko">field F</a>. </p><p>One considers first a vector space L that has the <a href="/facts/Cartesian_product/BfqBWzdL">Cartesian product</a> V × W {\displaystyle V\times W} as a <a href="/facts/Basis_(linear_algebra)/89IPoN6c">basis</a>. That is, the basis elements of L are the <a href="/facts/Ordered_pair/EInLVI0r">pairs</a> ( v , w ) {\displaystyle (v,w)} with v ∈ V {\displaystyle v\in V} and w ∈ W {\displaystyle w\in W} . To get such a vector space, one can define it as the vector space of the <a href="/facts/Function_(mathematics)/CdReEKCh">functions</a> V × W → F {\displaystyle V\times W\to F} that have a finite number of nonzero values and identifying ( v , w ) {\displaystyle (v,w)} with the function that takes the value 1 on ( v , w ) {\displaystyle (v,w)} and 0 otherwise. </p><p>Let R be the <a href="/facts/Linear_subspace/2wtKTgHL">linear subspace</a> of L that is spanned by the relations that the tensor product must satisfy. More precisely, R is <a href="/facts/Linear_span/vPjclEtb">spanned by</a> the elements of one of the forms: </p> ( v 1 + v 2 , w ) − ( v 1 , w ) − ( v 2 , w ) , ( v , w 1 + w 2 ) − ( v , w 1 ) − ( v , w 2 ) , ( s v , w ) − s ( v , w ) , ( v , s w ) − s ( v , w ) , {\displaystyle {\begin{aligned}(v_{1}+v_{2},w)&-(v_{1},w)-(v_{2},w),\\(v,w_{1}+w_{2})&-(v,w_{1})-(v,w_{2}),\\(sv,w)&-s(v,w),\\(v,sw)&-s(v,w),\end{aligned}}} <p>where v , v 1 , v 2 ∈ V {\displaystyle v,v_{1},v_{2}\in V} , w , w 1 , w 2 ∈ W {\displaystyle w,w_{1},w_{2}\in W} and s ∈ F {\displaystyle s\in F} . </p><p>Then, the tensor product is defined as the <a href="/facts/Quotient_space_(linear_algebra)/WmQ3lqdI">quotient space</a>: </p> V ⊗ W = L / R , {\displaystyle V\otimes W=L/R,} <p>and the image of ( v , w ) {\displaystyle (v,w)} in this quotient is denoted v ⊗ w {\displaystyle v\otimes w} . </p><p>It is straightforward to prove that the result of this construction satisfies the <a href="/facts/Universal_property/5XJBYMNz">universal property</a> considered below. (A very similar construction can be used to define the <a href="/facts/Tensor_product_of_modules/K3oaFIS4">tensor product of modules</a>.) </p> <h3>Universal property</h3> <p>In this section, the <a href="/facts/Universal_property/5XJBYMNz">universal property</a> satisfied by the tensor product is described. As for every universal property, two objects that satisfy the property are related by a unique <a href="/facts/Isomorphism/pj8KgkxU">isomorphism</a>. It follows that this is a (non-constructive) way to define the tensor product of two vector spaces. In this context, the preceding constructions of tensor products may be viewed as proofs of existence of the tensor product so defined. </p><p>A consequence of this approach is that every property of the tensor product can be deduced from the universal property, and that, in practice, one may forget the method that has been used to prove its existence. </p><p>The "universal-property definition" of the tensor product of two vector spaces is the following (recall that a <a href="/facts/Bilinear_map/LH1lHVeP">bilinear map</a> is a function that is <i>separately</i> <a href="/facts/Linear_map/Q1PBKNVV">linear</a> in each of its arguments): </p> The <i>tensor product</i> of two vector spaces V and W is a vector space denoted as V ⊗ W {\displaystyle V\otimes W} , together with a bilinear map ⊗ : ( v , w ) ↦ v ⊗ w {\displaystyle {\otimes }:(v,w)\mapsto v\otimes w} from V × W {\displaystyle V\times W} to V ⊗ W {\displaystyle V\otimes W} , such that, for every bilinear map h : V × W → Z {\displaystyle h:V\times W\to Z} , there is a <i>unique</i> linear map h ~ : V ⊗ W → Z {\displaystyle {\tilde {h}}:V\otimes W\to Z} , such that h = h ~ ∘ ⊗ {\displaystyle h={\tilde {h}}\circ {\otimes }} (that is, h ( v , w ) = h ~ ( v ⊗ w ) {\displaystyle h(v,w)={\tilde {h}}(v\otimes w)} for every v ∈ V {\displaystyle v\in V} and w ∈ W {\displaystyle w\in W} ). <h3>Linearly disjoint</h3> <p>Like the universal property above, the following characterization may also be used to determine whether or not a given vector space and given bilinear map form a tensor product.<a class="footnote-ref" id="fnref:1" href="#fn:1"><sup>1</sup></a> </p> <p>Theorem—Let X , Y {\displaystyle X,Y} , and Z {\displaystyle Z} be complex vector spaces and let T : X × Y → Z {\displaystyle T:X\times Y\to Z} be a bilinear map. Then ( Z , T ) {\displaystyle (Z,T)} is a tensor product of X {\displaystyle X} and Y {\displaystyle Y} if and only if<a class="footnote-ref" id="fnref:2" href="#fn:2"><sup>2</sup></a> the image of T {\displaystyle T} spans all of Z {\displaystyle Z} (that is, span T ( X × Y ) = Z {\displaystyle \operatorname {span} \;T(X\times Y)=Z} ), and also X {\displaystyle X} and Y {\displaystyle Y} are T {\displaystyle T} -linearly disjoint, which by definition means that for all positive integers n {\displaystyle n} and all elements x 1 , … , x n ∈ X {\displaystyle x_{1},\ldots ,x_{n}\in X} and y 1 , … , y n ∈ Y {\displaystyle y_{1},\ldots ,y_{n}\in Y} such that ∑ i = 1 n T ( x i , y i ) = 0 {\displaystyle \sum _{i=1}^{n}T\left(x_{i},y_{i}\right)=0} , </p> <ol><li>if all x 1 , … , x n {\displaystyle x_{1},\ldots ,x_{n}} are <a href="/facts/Linearly_independent/8JrHfIIa">linearly independent</a> then all y i {\displaystyle y_{i}} are 0 {\displaystyle 0} , and</li> <li>if all y 1 , … , y n {\displaystyle y_{1},\ldots ,y_{n}} are linearly independent then all x i {\displaystyle x_{i}} are 0 {\displaystyle 0} .</li></ol> <p>Equivalently, X {\displaystyle X} and Y {\displaystyle Y} are T {\displaystyle T} -linearly disjoint if and only if for all linearly independent sequences x 1 , … , x m {\displaystyle x_{1},\ldots ,x_{m}} in X {\displaystyle X} and all linearly independent sequences y 1 , … , y n {\displaystyle y_{1},\ldots ,y_{n}} in Y {\displaystyle Y} , the vectors { T ( x i , y j ) : 1 ≤ i ≤ m , 1 ≤ j ≤ n } {\displaystyle \left\{T\left(x_{i},y_{j}\right):1\leq i\leq m,1\leq j\leq n\right\}} are linearly independent. </p> <p>For example, it follows immediately that if X = C m {\displaystyle X=\mathbb {C} ^{m}} and Y = C n {\displaystyle Y=\mathbb {C} ^{n}} , where m {\displaystyle m} and n {\displaystyle n} are positive integers, then one may set Z = C m n {\displaystyle Z=\mathbb {C} ^{mn}} and define the bilinear map as T : C m × C n → C m n ( x , y ) = ( ( x 1 , … , x m ) , ( y 1 , … , y n ) ) ↦ ( x i y j ) j = 1 , … , n i = 1 , … , m {\displaystyle {\begin{aligned}T:\mathbb {C} ^{m}\times \mathbb {C} ^{n}&\to \mathbb {C} ^{mn}\\(x,y)=((x_{1},\ldots ,x_{m}),(y_{1},\ldots ,y_{n}))&\mapsto (x_{i}y_{j})_{\stackrel {i=1,\ldots ,m}{j=1,\ldots ,n}}\end{aligned}}} to form the tensor product of X {\displaystyle X} and Y {\displaystyle Y} .<a class="footnote-ref" id="fnref:3" href="#fn:3"><sup>3</sup></a> Often, this map T {\displaystyle T} is denoted by ⊗ {\displaystyle \,\otimes \,} so that x ⊗ y = T ( x , y ) . {\displaystyle x\otimes y=T(x,y).} </p><p>As another example, suppose that C S {\displaystyle \mathbb {C} ^{S}} is the vector space of all complex-valued functions on a set S {\displaystyle S} with addition and scalar multiplication defined pointwise (meaning that f + g {\displaystyle f+g} is the map s ↦ f ( s ) + g ( s ) {\displaystyle s\mapsto f(s)+g(s)} and c f {\displaystyle cf} is the map s ↦ c f ( s ) {\displaystyle s\mapsto cf(s)} ). Let S {\displaystyle S} and T {\displaystyle T} be any sets and for any f ∈ C S {\displaystyle f\in \mathbb {C} ^{S}} and g ∈ C T {\displaystyle g\in \mathbb {C} ^{T}} , let f ⊗ g ∈ C S × T {\displaystyle f\otimes g\in \mathbb {C} ^{S\times T}} denote the function defined by ( s , t ) ↦ f ( s ) g ( t ) {\displaystyle (s,t)\mapsto f(s)g(t)} . If X ⊆ C S {\displaystyle X\subseteq \mathbb {C} ^{S}} and Y ⊆ C T {\displaystyle Y\subseteq \mathbb {C} ^{T}} are vector subspaces then the vector subspace Z := span { f ⊗ g : f ∈ X , g ∈ Y } {\displaystyle Z:=\operatorname {span} \left\{f\otimes g:f\in X,g\in Y\right\}} of C S × T {\displaystyle \mathbb {C} ^{S\times T}} together with the bilinear map: X × Y → Z ( f , g ) ↦ f ⊗ g {\displaystyle {\begin{alignedat}{4}\;&&X\times Y&&\;\to \;&Z\\[0.3ex]&&(f,g)&&\;\mapsto \;&f\otimes g\\\end{alignedat}}} form a tensor product of X {\displaystyle X} and Y {\displaystyle Y} .<a class="footnote-ref" id="fnref:4" href="#fn:4"><sup>4</sup></a> </p> <h2 id="properties">Properties</h2> <h3>Dimension</h3> <p>If <i>V</i> and <i>W</i> are vector spaces of finite <a href="/facts/Dimension_(linear_algebra)/z8m02A3W">dimension</a>, then V ⊗ W {\displaystyle V\otimes W} is finite-dimensional, and its dimension is the product of the dimensions of <i>V</i> and <i>W</i>. </p><p>This results from the fact that a basis of V ⊗ W {\displaystyle V\otimes W} is formed by taking all tensor products of a basis element of <i>V</i> and a basis element of <i>W</i>. </p> <h3>Associativity</h3> <p>The tensor product is <a href="/facts/Associative/EQX8lV6r">associative</a> in the sense that, given three vector spaces U , V , W {\displaystyle U,V,W} , there is a canonical isomorphism: </p> ( U ⊗ V ) ⊗ W ≅ U ⊗ ( V ⊗ W ) , {\displaystyle (U\otimes V)\otimes W\cong U\otimes (V\otimes W),} <p>that maps ( u ⊗ v ) ⊗ w {\displaystyle (u\otimes v)\otimes w} to u ⊗ ( v ⊗ w ) {\displaystyle u\otimes (v\otimes w)} . </p><p>This allows omitting parentheses in the tensor product of more than two vector spaces or vectors. </p> <h3>Commutativity as vector space operation</h3> <p>The tensor product of two vector spaces V {\displaystyle V} and W {\displaystyle W} is <a href="/facts/Commutative/WaYxJLxd">commutative </a> in the sense that there is a canonical isomorphism: </p> V ⊗ W ≅ W ⊗ V , {\displaystyle V\otimes W\cong W\otimes V,} <p>that maps v ⊗ w {\displaystyle v\otimes w} to w ⊗ v {\displaystyle w\otimes v} . </p><p>On the other hand, even when V = W {\displaystyle V=W} , the tensor product of vectors is not commutative; that is v ⊗ w ≠ w ⊗ v {\displaystyle v\otimes w\neq w\otimes v} , in general. </p><p> The map x ⊗ y ↦ y ⊗ x {\displaystyle x\otimes y\mapsto y\otimes x} from V ⊗ V {\displaystyle V\otimes V} to itself induces a linear <a href="/facts/Automorphism/WnyxNFkQ">automorphism</a> that is called a braiding map. More generally and as usual (see <a href="/facts/Tensor_algebra/H7CKKryf">tensor algebra</a>), let V ⊗ n {\displaystyle V^{\otimes n}} denote the tensor product of n copies of the vector space V. For every <a href="/facts/Permutation/JSw9ukmv">permutation</a> s of the first n positive integers, the map: </p> x 1 ⊗ ⋯ ⊗ x n ↦ x s ( 1 ) ⊗ ⋯ ⊗ x s ( n ) {\displaystyle x_{1}\otimes \cdots \otimes x_{n}\mapsto x_{s(1)}\otimes \cdots \otimes x_{s(n)}} <p>induces a linear automorphism of V ⊗ n → V ⊗ n {\displaystyle V^{\otimes n}\to V^{\otimes n}} , which is called a braiding map. </p> <h2 id="tensor-product-of-linear-maps">Tensor product of linear maps</h2> <p class="note">"Tensor product of linear maps" redirects here. For the generalization for modules, see <a href="/facts/Tensor_product_of_modules/K3oaFIS4">Tensor product of modules § Tensor product of linear maps and a change of base ring</a>.</p> <p>Given a linear map f : U → V {\displaystyle f:U\to V} , and a vector space W, the <i>tensor product:</i> </p> f ⊗ W : U ⊗ W → V ⊗ W {\displaystyle f\otimes W:U\otimes W\to V\otimes W} <p>is the unique linear map such that: </p> ( f ⊗ W ) ( u ⊗ w ) = f ( u ) ⊗ w . {\displaystyle (f\otimes W)(u\otimes w)=f(u)\otimes w.} <p>The tensor product W ⊗ f {\displaystyle W\otimes f} is defined similarly. </p><p>Given two linear maps f : U → V {\displaystyle f:U\to V} and g : W → Z {\displaystyle g:W\to Z} , their tensor product: </p> f ⊗ g : U ⊗ W → V ⊗ Z {\displaystyle f\otimes g:U\otimes W\to V\otimes Z} <p>is the unique linear map that satisfies: </p> ( f ⊗ g ) ( u ⊗ w ) = f ( u ) ⊗ g ( w ) . {\displaystyle (f\otimes g)(u\otimes w)=f(u)\otimes g(w).} <p>One has: </p> f ⊗ g = ( f ⊗ Z ) ∘ ( U ⊗ g ) = ( V ⊗ g ) ∘ ( f ⊗ W ) . {\displaystyle f\otimes g=(f\otimes Z)\circ (U\otimes g)=(V\otimes g)\circ (f\otimes W).} <p>In terms of <a href="/facts/Category_theory/6zS3DXPa">category theory</a>, this means that the tensor product is a <a href="/facts/Bifunctor/l2u3rIBE">bifunctor</a> from the <a href="/facts/Category_(mathematics)/7WzxUThf">category</a> of vector spaces to itself.<a class="footnote-ref" id="fnref:5" href="#fn:5"><sup>5</sup></a> </p><p>If f and g are both <a href="/facts/Injective/5ES6tqf5">injective</a> or <a href="/facts/Surjective/KezhLpCb">surjective</a>, then the same is true for all above defined linear maps. In particular, the tensor product with a vector space is an <a href="/facts/Exact_functor/4tsQzwBp">exact functor</a>; this means that every <a href="/facts/Exact_sequence/leUJnwJb">exact sequence</a> is mapped to an exact sequence (<a href="/facts/Tensor_product_of_modules/K3oaFIS4">tensor products of modules</a> do not transform injections into injections, but they are <a href="/facts/Right_exact_functor/4tsQzwBp">right exact functors</a>). </p><p>By choosing bases of all vector spaces involved, the linear maps <i>f</i> and <i>g</i> can be represented by <a href="/facts/Matrix_(mathematics)/qa8FI4ko">matrices</a>. Then, depending on how the tensor v ⊗ w {\displaystyle v\otimes w} is vectorized, the matrix describing the tensor product f ⊗ g {\displaystyle f\otimes g} is the <a href="/facts/Kronecker_product/gABGx6I6">Kronecker product</a> of the two matrices. For example, if <i>V</i>, <i>X</i>, <i>W</i>, and <i>U</i> above are all two-dimensional and bases have been fixed for all of them, and <i>f</i> and <i>g</i> are given by the matrices: A = [ a 1 , 1 a 1 , 2 a 2 , 1 a 2 , 2 ] , B = [ b 1 , 1 b 1 , 2 b 2 , 1 b 2 , 2 ] , {\displaystyle A={\begin{bmatrix}a_{1,1}&a_{1,2}\\a_{2,1}&a_{2,2}\\\end{bmatrix}},\qquad B={\begin{bmatrix}b_{1,1}&b_{1,2}\\b_{2,1}&b_{2,2}\\\end{bmatrix}},} respectively, then the tensor product of these two matrices is: [ a 1 , 1 a 1 , 2 a 2 , 1 a 2 , 2 ] ⊗ [ b 1 , 1 b 1 , 2 b 2 , 1 b 2 , 2 ] = [ a 1 , 1 [ b 1 , 1 b 1 , 2 b 2 , 1 b 2 , 2 ] a 1 , 2 [ b 1 , 1 b 1 , 2 b 2 , 1 b 2 , 2 ] a 2 , 1 [ b 1 , 1 b 1 , 2 b 2 , 1 b 2 , 2 ] a 2 , 2 [ b 1 , 1 b 1 , 2 b 2 , 1 b 2 , 2 ] ] = [ a 1 , 1 b 1 , 1 a 1 , 1 b 1 , 2 a 1 , 2 b 1 , 1 a 1 , 2 b 1 , 2 a 1 , 1 b 2 , 1 a 1 , 1 b 2 , 2 a 1 , 2 b 2 , 1 a 1 , 2 b 2 , 2 a 2 , 1 b 1 , 1 a 2 , 1 b 1 , 2 a 2 , 2 b 1 , 1 a 2 , 2 b 1 , 2 a 2 , 1 b 2 , 1 a 2 , 1 b 2 , 2 a 2 , 2 b 2 , 1 a 2 , 2 b 2 , 2 ] . {\displaystyle {\begin{aligned}{\begin{bmatrix}a_{1,1}&a_{1,2}\\a_{2,1}&a_{2,2}\\\end{bmatrix}}\otimes {\begin{bmatrix}b_{1,1}&b_{1,2}\\b_{2,1}&b_{2,2}\\\end{bmatrix}}&={\begin{bmatrix}a_{1,1}{\begin{bmatrix}b_{1,1}&b_{1,2}\\b_{2,1}&b_{2,2}\\\end{bmatrix}}&a_{1,2}{\begin{bmatrix}b_{1,1}&b_{1,2}\\b_{2,1}&b_{2,2}\\\end{bmatrix}}\\[3pt]a_{2,1}{\begin{bmatrix}b_{1,1}&b_{1,2}\\b_{2,1}&b_{2,2}\\\end{bmatrix}}&a_{2,2}{\begin{bmatrix}b_{1,1}&b_{1,2}\\b_{2,1}&b_{2,2}\\\end{bmatrix}}\\\end{bmatrix}}\\&={\begin{bmatrix}a_{1,1}b_{1,1}&a_{1,1}b_{1,2}&a_{1,2}b_{1,1}&a_{1,2}b_{1,2}\\a_{1,1}b_{2,1}&a_{1,1}b_{2,2}&a_{1,2}b_{2,1}&a_{1,2}b_{2,2}\\a_{2,1}b_{1,1}&a_{2,1}b_{1,2}&a_{2,2}b_{1,1}&a_{2,2}b_{1,2}\\a_{2,1}b_{2,1}&a_{2,1}b_{2,2}&a_{2,2}b_{2,1}&a_{2,2}b_{2,2}\\\end{bmatrix}}.\end{aligned}}} </p><p>The resultant rank is at most 4, and thus the resultant dimension is 4. rank here denotes the <a href="/facts/Tensor_rank/2KuNm0MK">tensor rank</a> i.e. the number of requisite indices (while the <a href="/facts/Matrix_rank/Yp9b5LK3">matrix rank</a> counts the number of degrees of freedom in the resulting array). Tr A ⊗ B = Tr A × Tr B {\displaystyle \operatorname {Tr} A\otimes B=\operatorname {Tr} A\times \operatorname {Tr} B} . </p><p>A <a href="/facts/Dyadic_product/FoLJnrwR">dyadic product</a> is the special case of the tensor product between two vectors of the same dimension. </p> <h2 id="general-tensors">General tensors</h2> <p class="note">See also: <a href="/facts/Tensor/pNjFo5tV">Tensor</a></p> <p>For non-negative integers <i>r</i> and <i>s</i> a type ( r , s ) {\displaystyle (r,s)} <a href="/facts/Tensor/pNjFo5tV">tensor</a> on a vector space <i>V</i> is an element of: T s r ( V ) = V ⊗ ⋯ ⊗ V ⏟ r ⊗ V ∗ ⊗ ⋯ ⊗ V ∗ ⏟ s = V ⊗ r ⊗ ( V ∗ ) ⊗ s . {\displaystyle T_{s}^{r}(V)=\underbrace {V\otimes \cdots \otimes V} _{r}\otimes \underbrace {V^{*}\otimes \cdots \otimes V^{*}} _{s}=V^{\otimes r}\otimes \left(V^{*}\right)^{\otimes s}.} Here V ∗ {\displaystyle V^{*}} is the <a href="/facts/Dual_vector_space/4Uw4Knz1">dual vector space</a> (which consists of all <a href="/facts/Linear_map/Q1PBKNVV">linear maps</a> <i>f</i> from <i>V</i> to the ground field <i>K</i>). </p><p>There is a product map, called the (tensor) product of tensors:<a class="footnote-ref" id="fnref:6" href="#fn:6"><sup>6</sup></a> T s r ( V ) ⊗ K T s ′ r ′ ( V ) → T s + s ′ r + r ′ ( V ) . {\displaystyle T_{s}^{r}(V)\otimes _{K}T_{s'}^{r'}(V)\to T_{s+s'}^{r+r'}(V).} </p><p>It is defined by grouping all occurring "factors" <i>V</i> together: writing v i {\displaystyle v_{i}} for an element of <i>V</i> and f i {\displaystyle f_{i}} for an element of the dual space: ( v 1 ⊗ f 1 ) ⊗ ( v 1 ′ ) = v 1 ⊗ v 1 ′ ⊗ f 1 . {\displaystyle (v_{1}\otimes f_{1})\otimes (v'_{1})=v_{1}\otimes v'_{1}\otimes f_{1}.} </p><p>If <i>V</i> is finite dimensional, then picking a basis of <i>V</i> and the corresponding <a href="/facts/Dual_basis/3S14kq2A">dual basis</a> of V ∗ {\displaystyle V^{*}} naturally induces a basis of T s r ( V ) {\displaystyle T_{s}^{r}(V)} (this basis is described in the <a href="/facts/Kronecker_product/gABGx6I6">article on Kronecker products</a>). In terms of these bases, the <a href="/facts/Coordinate_vector/TEcLqS40">components</a> of a (tensor) product of two (or more) <a href="/facts/Tensor/pNjFo5tV">tensors</a> can be computed. For example, if <i>F</i> and <i>G</i> are two <a href="/facts/Covariance_and_contravariance_of_vectors/s3HGMkhz">covariant</a> tensors of orders <i>m</i> and <i>n</i> respectively (i.e. F ∈ T m 0 {\displaystyle F\in T_{m}^{0}} and G ∈ T n 0 {\displaystyle G\in T_{n}^{0}} ), then the components of their tensor product are given by:<a class="footnote-ref" id="fnref:7" href="#fn:7"><sup>7</sup></a> ( F ⊗ G ) i 1 i 2 ⋯ i m + n = F i 1 i 2 ⋯ i m G i m + 1 i m + 2 i m + 3 ⋯ i m + n . {\displaystyle (F\otimes G)_{i_{1}i_{2}\cdots i_{m+n}}=F_{i_{1}i_{2}\cdots i_{m}}G_{i_{m+1}i_{m+2}i_{m+3}\cdots i_{m+n}}.} </p><p>Thus, the components of the tensor product of two tensors are the ordinary product of the components of each tensor. Another example: let U be a tensor of type (1, 1) with components U β α {\displaystyle U_{\beta }^{\alpha }} , and let V be a tensor of type ( 1 , 0 ) {\displaystyle (1,0)} with components V γ {\displaystyle V^{\gamma }} . Then: ( U ⊗ V ) α β γ = U α β V γ {\displaystyle \left(U\otimes V\right)^{\alpha }{}_{\beta }{}^{\gamma }=U^{\alpha }{}_{\beta }V^{\gamma }} and: ( V ⊗ U ) μ ν σ = V μ U ν σ . {\displaystyle (V\otimes U)^{\mu \nu }{}_{\sigma }=V^{\mu }U^{\nu }{}_{\sigma }.} </p><p>Tensors equipped with their product operation form an <a href="/facts/Algebra_over_a_field/8T8ulWJb">algebra</a>, called the <a href="/facts/Tensor_algebra/H7CKKryf">tensor algebra</a>. </p> <h3>Evaluation map and tensor contraction</h3> <p>For tensors of type (1, 1) there is a canonical evaluation map: V ⊗ V ∗ → K {\displaystyle V\otimes V^{*}\to K} defined by its action on pure tensors: v ⊗ f ↦ f ( v ) . {\displaystyle v\otimes f\mapsto f(v).} </p><p>More generally, for tensors of type ( r , s ) {\displaystyle (r,s)} , with <i>r</i>, <i>s</i> > 0, there is a map, called <a href="/facts/Tensor_contraction/fqz0zvWp">tensor contraction</a>: T s r ( V ) → T s − 1 r − 1 ( V ) . {\displaystyle T_{s}^{r}(V)\to T_{s-1}^{r-1}(V).} (The copies of V {\displaystyle V} and V ∗ {\displaystyle V^{*}} on which this map is to be applied must be specified.) </p><p>On the other hand, if V {\displaystyle V} is finite-dimensional, there is a canonical map in the other direction (called the coevaluation map): { K → V ⊗ V ∗ λ ↦ ∑ i λ v i ⊗ v i ∗ {\displaystyle {\begin{cases}K\to V\otimes V^{*}\\\lambda \mapsto \sum _{i}\lambda v_{i}\otimes v_{i}^{*}\end{cases}}} where v 1 , … , v n {\displaystyle v_{1},\ldots ,v_{n}} is any basis of V {\displaystyle V} , and v i ∗ {\displaystyle v_{i}^{*}} is its <a href="/facts/Dual_basis/3S14kq2A">dual basis</a>. This map does not depend on the choice of basis.<a class="footnote-ref" id="fnref:8" href="#fn:8"><sup>8</sup></a> </p><p>The interplay of evaluation and coevaluation can be used to characterize finite-dimensional vector spaces without referring to bases.<a class="footnote-ref" id="fnref:9" href="#fn:9"><sup>9</sup></a> </p> <h3>Adjoint representation</h3> <p>The tensor product T s r ( V ) {\displaystyle T_{s}^{r}(V)} may be naturally viewed as a module for the <a href="/facts/Lie_algebra/n2gAUkNq">Lie algebra</a> E n d ( V ) {\displaystyle \mathrm {End} (V)} by means of the diagonal action: for simplicity let us assume r = s = 1 {\displaystyle r=s=1} , then, for each u ∈ E n d ( V ) {\displaystyle u\in \mathrm {End} (V)} , u ( a ⊗ b ) = u ( a ) ⊗ b − a ⊗ u ∗ ( b ) , {\displaystyle u(a\otimes b)=u(a)\otimes b-a\otimes u^{*}(b),} where u ∗ ∈ E n d ( V ∗ ) {\displaystyle u^{*}\in \mathrm {End} \left(V^{*}\right)} is the <a href="/facts/Transpose/8wmsagGS">transpose</a> of <i>u</i>, that is, in terms of the obvious pairing on V ⊗ V ∗ {\displaystyle V\otimes V^{*}} , ⟨ u ( a ) , b ⟩ = ⟨ a , u ∗ ( b ) ⟩ . {\displaystyle \langle u(a),b\rangle =\langle a,u^{*}(b)\rangle .} </p><p>There is a canonical isomorphism T 1 1 ( V ) → E n d ( V ) {\displaystyle T_{1}^{1}(V)\to \mathrm {End} (V)} given by: ( a ⊗ b ) ( x ) = ⟨ x , b ⟩ a . {\displaystyle (a\otimes b)(x)=\langle x,b\rangle a.} </p><p>Under this isomorphism, every <i>u</i> in E n d ( V ) {\displaystyle \mathrm {End} (V)} may be first viewed as an endomorphism of T 1 1 ( V ) {\displaystyle T_{1}^{1}(V)} and then viewed as an endomorphism of E n d ( V ) {\displaystyle \mathrm {End} (V)} . In fact it is the <a href="/facts/Adjoint_representation_of_a_Lie_algebra/bWG4c2su">adjoint representation</a> ad(<i>u</i>) of E n d ( V ) {\displaystyle \mathrm {End} (V)} . </p> <h2 id="linear-maps-as-tensors">Linear maps as tensors</h2> <p>Given two finite dimensional vector spaces <i>U</i>, <i>V</i> over the same field <i>K</i>, denote the <a href="/facts/Dual_space/4Uw4Knz1">dual space</a> of <i>U</i> as <i>U*</i>, and the <i>K</i>-vector space of all linear maps from <i>U</i> to <i>V</i> as Hom(<i>U</i>,<i>V</i>). There is an isomorphism: U ∗ ⊗ V ≅ H o m ( U , V ) , {\displaystyle U^{*}\otimes V\cong \mathrm {Hom} (U,V),} defined by an action of the pure tensor f ⊗ v ∈ U ∗ ⊗ V {\displaystyle f\otimes v\in U^{*}\otimes V} on an element of U {\displaystyle U} , ( f ⊗ v ) ( u ) = f ( u ) v . {\displaystyle (f\otimes v)(u)=f(u)v.} </p><p>Its "inverse" can be defined using a basis { u i } {\displaystyle \{u_{i}\}} and its dual basis { u i ∗ } {\displaystyle \{u_{i}^{*}\}} as in the section "Evaluation map and tensor contraction" above: { H o m ( U , V ) → U ∗ ⊗ V F ↦ ∑ i u i ∗ ⊗ F ( u i ) . {\displaystyle {\begin{cases}\mathrm {Hom} (U,V)\to U^{*}\otimes V\\F\mapsto \sum _{i}u_{i}^{*}\otimes F(u_{i}).\end{cases}}} </p><p>This result implies: dim ( U ⊗ V ) = dim ( U ) dim ( V ) , {\displaystyle \dim(U\otimes V)=\dim(U)\dim(V),} which automatically gives the important fact that { u i ⊗ v j } {\displaystyle \{u_{i}\otimes v_{j}\}} forms a basis of U ⊗ V {\displaystyle U\otimes V} where { u i } , { v j } {\displaystyle \{u_{i}\},\{v_{j}\}} are bases of <i>U</i> and <i>V</i>. </p><p>Furthermore, given three vector spaces <i>U</i>, <i>V</i>, <i>W</i> the tensor product is linked to the vector space of <i>all</i> linear maps, as follows: H o m ( U ⊗ V , W ) ≅ H o m ( U , H o m ( V , W ) ) . {\displaystyle \mathrm {Hom} (U\otimes V,W)\cong \mathrm {Hom} (U,\mathrm {Hom} (V,W)).} This is an example of <a href="/facts/Adjoint_functor/bJ1P7AN2">adjoint functors</a>: the tensor product is "left adjoint" to Hom. </p> <h2 id="tensor-products-of-modules-over-a-ring">Tensor products of modules over a ring</h2> <p class="note">Main article: <a href="/facts/Tensor_product_of_modules/K3oaFIS4">Tensor product of modules</a></p> <p>The tensor product of two <a href="/facts/Module_(mathematics)/jP0LIhFL">modules</a> <i>A</i> and <i>B</i> over a <i><a href="/facts/Commutative_ring/QS1r2ovR">commutative</a></i> <a href="/facts/Ring_(mathematics)/JnCUXz63">ring</a> <i>R</i> is defined in exactly the same way as the tensor product of vector spaces over a field: A ⊗ R B := F ( A × B ) / G , {\displaystyle A\otimes _{R}B:=F(A\times B)/G,} where now F ( A × B ) {\displaystyle F(A\times B)} is the <a href="/facts/Free_module/o6rC6yoJ">free <i>R</i>-module</a> generated by the cartesian product and <i>G</i> is the <i>R</i>-module generated by <a href="/facts/Tensor_product_of_modules/K3oaFIS4">these relations</a>. </p><p>More generally, the tensor product can be defined even if the ring is <a href="/facts/Noncommutative_ring/3jmll7q7">non-commutative</a>. In this case <i>A</i> has to be a right-<i>R</i>-module and <i>B</i> is a left-<i>R</i>-module, and instead of the last two relations above, the relation: ( a r , b ) ∼ ( a , r b ) {\displaystyle (ar,b)\sim (a,rb)} is imposed. If <i>R</i> is non-commutative, this is no longer an <i>R</i>-module, but just an <a href="/facts/Abelian_group/FfWwmx4R">abelian group</a>. </p><p>The universal property also carries over, slightly modified: the map φ : A × B → A ⊗ R B {\displaystyle \varphi :A\times B\to A\otimes _{R}B} defined by ( a , b ) ↦ a ⊗ b {\displaystyle (a,b)\mapsto a\otimes b} is a <a href="/facts/Tensor_product_of_modules/K3oaFIS4">middle linear map</a> (referred to as "the canonical middle linear map"<a class="footnote-ref" id="fnref:10" href="#fn:10"><sup>10</sup></a>); that is, it satisfies:<a class="footnote-ref" id="fnref:11" href="#fn:11"><sup>11</sup></a> φ ( a + a ′ , b ) = φ ( a , b ) + φ ( a ′ , b ) φ ( a , b + b ′ ) = φ ( a , b ) + φ ( a , b ′ ) φ ( a r , b ) = φ ( a , r b ) {\displaystyle {\begin{aligned}\varphi (a+a',b)&=\varphi (a,b)+\varphi (a',b)\\\varphi (a,b+b')&=\varphi (a,b)+\varphi (a,b')\\\varphi (ar,b)&=\varphi (a,rb)\end{aligned}}} </p><p>The first two properties make <i>φ</i> a bilinear map of the <a href="/facts/Abelian_group/FfWwmx4R">abelian group</a> A × B {\displaystyle A\times B} . For any middle linear map ψ {\displaystyle \psi } of A × B {\displaystyle A\times B} , a unique group homomorphism <i>f</i> of A ⊗ R B {\displaystyle A\otimes _{R}B} satisfies ψ = f ∘ φ {\displaystyle \psi =f\circ \varphi } , and this property determines φ {\displaystyle \varphi } within group isomorphism. See the <a href="/facts/Tensor_product_of_modules/K3oaFIS4">main article</a> for details. </p> <h3>Tensor product of modules over a non-commutative ring</h3> <p>Let <i>A</i> be a right <i>R</i>-module and <i>B</i> be a left <i>R</i>-module. Then the tensor product of <i>A</i> and <i>B</i> is an abelian group defined by: A ⊗ R B := F ( A × B ) / G {\displaystyle A\otimes _{R}B:=F(A\times B)/G} where F ( A × B ) {\displaystyle F(A\times B)} is a <a href="/facts/Free_abelian_group/oaSRGmpY">free abelian group</a> over A × B {\displaystyle A\times B} and G is the subgroup of F ( A × B ) {\displaystyle F(A\times B)} generated by relations: ∀ a , a 1 , a 2 ∈ A , ∀ b , b 1 , b 2 ∈ B , for all r ∈ R : ( a 1 , b ) + ( a 2 , b ) − ( a 1 + a 2 , b ) , ( a , b 1 ) + ( a , b 2 ) − ( a , b 1 + b 2 ) , ( a r , b ) − ( a , r b ) . {\displaystyle {\begin{aligned}&\forall a,a_{1},a_{2}\in A,\forall b,b_{1},b_{2}\in B,{\text{ for all }}r\in R:\\&(a_{1},b)+(a_{2},b)-(a_{1}+a_{2},b),\\&(a,b_{1})+(a,b_{2})-(a,b_{1}+b_{2}),\\&(ar,b)-(a,rb).\\\end{aligned}}} </p><p>The universal property can be stated as follows. Let <i>G</i> be an abelian group with a map q : A × B → G {\displaystyle q:A\times B\to G} that is bilinear, in the sense that: q ( a 1 + a 2 , b ) = q ( a 1 , b ) + q ( a 2 , b ) , q ( a , b 1 + b 2 ) = q ( a , b 1 ) + q ( a , b 2 ) , q ( a r , b ) = q ( a , r b ) . {\displaystyle {\begin{aligned}q(a_{1}+a_{2},b)&=q(a_{1},b)+q(a_{2},b),\\q(a,b_{1}+b_{2})&=q(a,b_{1})+q(a,b_{2}),\\q(ar,b)&=q(a,rb).\end{aligned}}} </p><p>Then there is a unique map q ¯ : A ⊗ B → G {\displaystyle {\overline {q}}:A\otimes B\to G} such that q ¯ ( a ⊗ b ) = q ( a , b ) {\displaystyle {\overline {q}}(a\otimes b)=q(a,b)} for all a ∈ A {\displaystyle a\in A} and b ∈ B {\displaystyle b\in B} . </p><p>Furthermore, we can give A ⊗ R B {\displaystyle A\otimes _{R}B} a module structure under some extra conditions: </p> <ol><li>If <i>A</i> is a (<i>S</i>,<i>R</i>)-bimodule, then A ⊗ R B {\displaystyle A\otimes _{R}B} is a left <i>S</i>-module, where s ( a ⊗ b ) := ( s a ) ⊗ b {\displaystyle s(a\otimes b):=(sa)\otimes b} .</li> <li>If <i>B</i> is a (<i>R</i>,<i>S</i>)-bimodule, then A ⊗ R B {\displaystyle A\otimes _{R}B} is a right <i>S</i>-module, where ( a ⊗ b ) s := a ⊗ ( b s ) {\displaystyle (a\otimes b)s:=a\otimes (bs)} .</li> <li>If <i>A</i> is a (<i>S</i>,<i>R</i>)-bimodule and <i>B</i> is a (<i>R</i>,<i>T</i>)-bimodule, then A ⊗ R B {\displaystyle A\otimes _{R}B} is a (<i>S</i>,<i>T</i>)-bimodule, where the left and right actions are defined in the same way as the previous two examples.</li> <li>If <i>R</i> is a commutative ring, then <i>A</i> and <i>B</i> are (<i>R</i>,<i>R</i>)-bimodules where r a := a r {\displaystyle ra:=ar} and b r := r b {\displaystyle br:=rb} . By 3), we can conclude A ⊗ R B {\displaystyle A\otimes _{R}B} is a (<i>R</i>,<i>R</i>)-bimodule.</li></ol> <h3>Computing the tensor product</h3> <p>For vector spaces, the tensor product V ⊗ W {\displaystyle V\otimes W} is quickly computed since bases of <i>V</i> of <i>W</i> immediately determine a basis of V ⊗ W {\displaystyle V\otimes W} , as was mentioned above. For modules over a general (commutative) ring, not every module is free. For example, Z/<i>n</i>Z is not a free abelian group (Z-module). The tensor product with Z/<i>n</i>Z is given by: M ⊗ Z Z / n Z = M / n M . {\displaystyle M\otimes _{\mathbf {Z} }\mathbf {Z} /n\mathbf {Z} =M/nM.} </p><p>More generally, given a <a href="/facts/Presentation_of_a_module/GxxKGm4L">presentation</a> of some <i>R</i>-module <i>M</i>, that is, a number of generators m i ∈ M , i ∈ I {\displaystyle m_{i}\in M,i\in I} together with relations: ∑ j ∈ J a j i m i = 0 , a i j ∈ R , {\displaystyle \sum _{j\in J}a_{ji}m_{i}=0,\qquad a_{ij}\in R,} the tensor product can be computed as the following <a href="/facts/Cokernel/vYFQRPoh">cokernel</a>: M ⊗ R N = coker ( N J → N I ) {\displaystyle M\otimes _{R}N=\operatorname {coker} \left(N^{J}\to N^{I}\right)} </p><p>Here N J = ⊕ j ∈ J N {\displaystyle N^{J}=\oplus _{j\in J}N} , and the map N J → N I {\displaystyle N^{J}\to N^{I}} is determined by sending some n ∈ N {\displaystyle n\in N} in the <i>j</i>th copy of N J {\displaystyle N^{J}} to a i j n {\displaystyle a_{ij}n} (in N I {\displaystyle N^{I}} ). Colloquially, this may be rephrased by saying that a presentation of <i>M</i> gives rise to a presentation of M ⊗ R N {\displaystyle M\otimes _{R}N} . This is referred to by saying that the tensor product is a <a href="/facts/Right_exact_functor/4tsQzwBp">right exact functor</a>. It is not in general left exact, that is, given an injective map of <i>R</i>-modules M 1 → M 2 {\displaystyle M_{1}\to M_{2}} , the tensor product: M 1 ⊗ R N → M 2 ⊗ R N {\displaystyle M_{1}\otimes _{R}N\to M_{2}\otimes _{R}N} is not usually injective. For example, tensoring the (injective) map given by multiplication with <i>n</i>, <i>n</i> : Z → Z with Z/<i>n</i>Z yields the zero map 0 : Z/<i>n</i>Z → Z/<i>n</i>Z, which is not injective. Higher <a href="/facts/Tor_functor/oFn2L3gd">Tor functors</a> measure the defect of the tensor product being not left exact. All higher Tor functors are assembled in the <a href="/facts/Derived_tensor_product/UHm1UG6z">derived tensor product</a>. </p> <h2 id="tensor-product-of-algebras">Tensor product of algebras</h2> <p class="note">Main article: <a href="/facts/Tensor_product_of_algebras/TdWszpzf">Tensor product of algebras</a></p> <p>Let <i>R</i> be a commutative ring. The tensor product of <i>R</i>-modules applies, in particular, if <i>A</i> and <i>B</i> are <a href="/facts/Algebra_(ring_theory)/8T8ulWJb"><i>R</i>-algebras</a>. In this case, the tensor product A ⊗ R B {\displaystyle A\otimes _{R}B} is an <i>R</i>-algebra itself by putting: ( a 1 ⊗ b 1 ) ⋅ ( a 2 ⊗ b 2 ) = ( a 1 ⋅ a 2 ) ⊗ ( b 1 ⋅ b 2 ) . {\displaystyle (a_{1}\otimes b_{1})\cdot (a_{2}\otimes b_{2})=(a_{1}\cdot a_{2})\otimes (b_{1}\cdot b_{2}).} For example: R [ x ] ⊗ R R [ y ] ≅ R [ x , y ] . {\displaystyle R[x]\otimes _{R}R[y]\cong R[x,y].} </p><p>A particular example is when <i>A</i> and <i>B</i> are fields containing a common subfield <i>R</i>. The <a href="/facts/Tensor_product_of_fields/FklJQo5j">tensor product of fields</a> is closely related to <a href="/facts/Galois_theory/54wscJfj">Galois theory</a>: if, say, <i>A</i> = <i>R</i>[<i>x</i>] / <i>f</i>(<i>x</i>), where <i>f</i> is some <a href="/facts/Irreducible_polynomial/PTeuKm4W">irreducible polynomial</a> with coefficients in <i>R</i>, the tensor product can be calculated as: A ⊗ R B ≅ B [ x ] / f ( x ) {\displaystyle A\otimes _{R}B\cong B[x]/f(x)} where now <i>f</i> is interpreted as the same polynomial, but with its coefficients regarded as elements of <i>B</i>. In the larger field <i>B</i>, the polynomial may become reducible, which brings in Galois theory. For example, if <i>A</i> = <i>B</i> is a <a href="/facts/Galois_extension/52fzNQbv">Galois extension</a> of <i>R</i>, then: A ⊗ R A ≅ A [ x ] / f ( x ) {\displaystyle A\otimes _{R}A\cong A[x]/f(x)} is isomorphic (as an <i>A</i>-algebra) to the A deg ( f ) {\displaystyle A^{\operatorname {deg} (f)}} . </p> <h2 id="eigenconfigurations-of-tensors">Eigenconfigurations of tensors</h2> <p>Square <a href="/facts/Matrix_(mathematics)/qa8FI4ko">matrices</a> A {\displaystyle A} with entries in a <a href="/facts/Field_(mathematics)/xAjAS4ko">field</a> K {\displaystyle K} represent <a href="/facts/Linear_map/Q1PBKNVV">linear maps</a> of <a href="/facts/Vector_space/M1pxMLx2">vector spaces</a>, say K n → K n {\displaystyle K^{n}\to K^{n}} , and thus linear maps ψ : P n − 1 → P n − 1 {\displaystyle \psi :\mathbb {P} ^{n-1}\to \mathbb {P} ^{n-1}} of <a href="/facts/Projective_spaces/7MSpA9cM">projective spaces</a> over K {\displaystyle K} . If A {\displaystyle A} is <a href="/facts/Invertible_matrix/fPqXk3V8">nonsingular</a> then ψ {\displaystyle \psi } is <a href="/facts/Well-defined/GAc0XbWz">well-defined</a> everywhere, and the <a href="/facts/Eigenvalues_and_eigenvectors/8TjEoT8u">eigenvectors</a> of A {\displaystyle A} correspond to the fixed points of ψ {\displaystyle \psi } . The <i>eigenconfiguration</i> of A {\displaystyle A} consists of n {\displaystyle n} points in P n − 1 {\displaystyle \mathbb {P} ^{n-1}} , provided A {\displaystyle A} is generic and K {\displaystyle K} is <a href="/facts/Algebraically_closed_field/HuVQClok">algebraically closed</a>. The fixed points of nonlinear maps are the eigenvectors of tensors. Let A = ( a i 1 i 2 ⋯ i d ) {\displaystyle A=(a_{i_{1}i_{2}\cdots i_{d}})} be a d {\displaystyle d} -dimensional tensor of format n × n × ⋯ × n {\displaystyle n\times n\times \cdots \times n} with entries ( a i 1 i 2 ⋯ i d ) {\displaystyle (a_{i_{1}i_{2}\cdots i_{d}})} lying in an algebraically closed field K {\displaystyle K} of <a href="/facts/Characteristic_(algebra)/D2VzcQaG">characteristic</a> zero. Such a tensor A ∈ ( K n ) ⊗ d {\displaystyle A\in (K^{n})^{\otimes d}} defines <a href="/facts/Morphism_of_algebraic_varieties/5jBzIgps">polynomial maps</a> K n → K n {\displaystyle K^{n}\to K^{n}} and P n − 1 → P n − 1 {\displaystyle \mathbb {P} ^{n-1}\to \mathbb {P} ^{n-1}} with coordinates: ψ i ( x 1 , … , x n ) = ∑ j 2 = 1 n ∑ j 3 = 1 n ⋯ ∑ j d = 1 n a i j 2 j 3 ⋯ j d x j 2 x j 3 ⋯ x j d for i = 1 , … , n {\displaystyle \psi _{i}(x_{1},\ldots ,x_{n})=\sum _{j_{2}=1}^{n}\sum _{j_{3}=1}^{n}\cdots \sum _{j_{d}=1}^{n}a_{ij_{2}j_{3}\cdots j_{d}}x_{j_{2}}x_{j_{3}}\cdots x_{j_{d}}\;\;{\mbox{for }}i=1,\ldots ,n} </p><p>Thus each of the n {\displaystyle n} coordinates of ψ {\displaystyle \psi } is a <a href="/facts/Homogeneous_polynomial/WUHL7Kct">homogeneous polynomial</a> ψ i {\displaystyle \psi _{i}} of degree d − 1 {\displaystyle d-1} in x = ( x 1 , … , x n ) {\displaystyle \mathbf {x} =\left(x_{1},\ldots ,x_{n}\right)} . The eigenvectors of A {\displaystyle A} are the solutions of the constraint: rank ( x 1 x 2 ⋯ x n ψ 1 ( x ) ψ 2 ( x ) ⋯ ψ n ( x ) ) ≤ 1 {\displaystyle {\mbox{rank}}{\begin{pmatrix}x_{1}&x_{2}&\cdots &x_{n}\\\psi _{1}(\mathbf {x} )&\psi _{2}(\mathbf {x} )&\cdots &\psi _{n}(\mathbf {x} )\end{pmatrix}}\leq 1} and the eigenconfiguration is given by the <a href="/facts/Algebraic_variety/Vk7f97vn">variety</a> of the 2 × 2 {\displaystyle 2\times 2} <a href="/facts/Minor_(linear_algebra)/CMudF5Eh">minors</a> of this matrix.<a class="footnote-ref" id="fnref:12" href="#fn:12"><sup>12</sup></a> </p> <h2 id="other-examples-of-tensor-products">Other examples of tensor products</h2> <h3>Topological tensor products</h3> <p class="note">Main articles: <a href="/facts/Topological_tensor_product/eDICDJ2Y">Topological tensor product</a> and <a href="/facts/Tensor_product_of_Hilbert_spaces/GSavLJfU">Tensor product of Hilbert spaces</a></p> <p><a href="/facts/Hilbert_space/aDFMh4lV">Hilbert spaces</a> generalize finite-dimensional vector spaces to arbitrary dimensions. There is <a href="/facts/Tensor_product_of_Hilbert_spaces/GSavLJfU">an analogous operation</a>, also called the "tensor product," that makes Hilbert spaces a <a href="/facts/Symmetric_monoidal_category/AxhumNny">symmetric monoidal category</a>. It is essentially constructed as the <a href="/facts/Complete_metric_space/lsckmU8G">metric space completion</a> of the algebraic tensor product discussed above. However, it does not satisfy the obvious analogue of the universal property defining tensor products;<a class="footnote-ref" id="fnref:13" href="#fn:13"><sup>13</sup></a> the morphisms for that property must be restricted to <a href="/facts/Hilbert%25E2%2580%2593Schmidt_operator/Xhy17yn6">Hilbert–Schmidt operators</a>.<a class="footnote-ref" id="fnref:14" href="#fn:14"><sup>14</sup></a> </p><p>In situations where the imposition of an inner product is inappropriate, one can still attempt to complete the algebraic tensor product, as a <a href="/facts/Topological_tensor_product/eDICDJ2Y">topological tensor product</a>. However, such a construction is no longer uniquely specified: in many cases, there are multiple natural topologies on the algebraic tensor product. </p> <h3>Tensor product of graded vector spaces</h3> <p class="note">Main article: <a href="/facts/Graded_vector_space/Ff1maywV">Graded vector space § Operations on graded vector spaces</a></p> <p>Some vector spaces can be decomposed into <a href="/facts/Direct_sum/x3ljVsP6">direct sums</a> of subspaces. In such cases, the tensor product of two spaces can be decomposed into sums of products of the subspaces (in analogy to the way that multiplication distributes over addition). </p> <h3>Tensor product of representations</h3> <p class="note">Main article: <a href="/facts/Tensor_product_of_representations/RXEV8te8">Tensor product of representations</a></p> <p>Vector spaces endowed with an additional multiplicative structure are called <a href="/facts/Algebra_over_a_field/8T8ulWJb">algebras</a>. The tensor product of such algebras is described by the <a href="/facts/Littlewood%25E2%2580%2593Richardson_rule/8NSx5K36">Littlewood–Richardson rule</a>. </p> <h3>Tensor product of quadratic forms</h3> <p class="note">Main article: <a href="/facts/Tensor_product_of_quadratic_forms/T58SgDK6">Tensor product of quadratic forms</a></p> <h3>Tensor product of multilinear forms</h3> <p>Given two <a href="/facts/Multilinear_form/Rk420AcK">multilinear forms</a> f ( x 1 , … , x k ) {\displaystyle f(x_{1},\dots ,x_{k})} and g ( x 1 , … , x m ) {\displaystyle g(x_{1},\dots ,x_{m})} on a vector space V {\displaystyle V} over the field K {\displaystyle K} their tensor product is the multilinear form: ( f ⊗ g ) ( x 1 , … , x k + m ) = f ( x 1 , … , x k ) g ( x k + 1 , … , x k + m ) . {\displaystyle (f\otimes g)(x_{1},\dots ,x_{k+m})=f(x_{1},\dots ,x_{k})g(x_{k+1},\dots ,x_{k+m}).} <a class="footnote-ref" id="fnref:15" href="#fn:15"><sup>15</sup></a> </p><p>This is a special case of the product of tensors if they are seen as multilinear maps (see also <a href="/facts/Tensor/pNjFo5tV">tensors as multilinear maps</a>). Thus the components of the tensor product of multilinear forms can be computed by the <a href="/facts/Kronecker_product/gABGx6I6">Kronecker product</a>. </p> <h3>Tensor product of sheaves of modules</h3> <p class="note">Main article: <a href="/facts/Sheaf_of_modules/grSMBXho">Sheaf of modules</a></p> <h3>Tensor product of line bundles</h3> <p class="note">Main article: <a href="/facts/Vector_bundle/cXACAa1I">Vector bundle § Operations on vector bundles</a></p> <p class="note">See also: <a href="/facts/Tensor_product_bundle/uqw5NYgu">Tensor product bundle</a></p> <h3>Tensor product of fields</h3> <p class="note">Main article: <a href="/facts/Tensor_product_of_fields/FklJQo5j">Tensor product of fields</a></p> <h3>Tensor product of graphs</h3> <p class="note">Main article: <a href="/facts/Tensor_product_of_graphs/pF1UjCvq">Tensor product of graphs</a></p> <p>It should be mentioned that, though called "tensor product", this is not a tensor product of graphs in the above sense; actually it is the <a href="/facts/Product_(category_theory)/evFecz0t">category-theoretic product</a> in the category of graphs and <a href="/facts/Graph_homomorphism/XSljQ25z">graph homomorphisms</a>. However it is actually the <a href="/facts/Kronecker_product/gABGx6I6">Kronecker tensor product</a> of the <a href="/facts/Adjacency_matrix/jYXNhYGs">adjacency matrices</a> of the graphs. Compare also the section Tensor product of linear maps above. </p> <h3>Monoidal categories</h3> <p>The most general setting for the tensor product is the <a href="/facts/Monoidal_category/F2l5oKMD">monoidal category</a>. It captures the algebraic essence of tensoring, without making any specific reference to what is being tensored. Thus, all tensor products can be expressed as an application of the monoidal category to some particular setting, acting on some particular objects. </p> <h2 id="quotient-algebras">Quotient algebras</h2> <p>A number of important subspaces of the <a href="/facts/Tensor_algebra/H7CKKryf">tensor algebra</a> can be constructed as <a href="/facts/Quotient_space_(linear_algebra)/WmQ3lqdI">quotients</a>: these include the <a href="/facts/Exterior_algebra/rdN4TvpR">exterior algebra</a>, the <a href="/facts/Symmetric_algebra/A8FthY0N">symmetric algebra</a>, the <a href="/facts/Clifford_algebra/qnqSCz5h">Clifford algebra</a>, the <a href="/facts/Weyl_algebra/PADSD0k9">Weyl algebra</a>, and the <a href="/facts/Universal_enveloping_algebra/VJ3Syz4W">universal enveloping algebra</a> in general. </p><p>The exterior algebra is constructed from the <a href="/facts/Exterior_product/rdN4TvpR">exterior product</a>. Given a vector space <i>V</i>, the exterior product V ∧ V {\displaystyle V\wedge V} is defined as: V ∧ V := V ⊗ V / { v ⊗ v ∣ v ∈ V } . {\displaystyle V\wedge V:=V\otimes V{\big /}\{v\otimes v\mid v\in V\}.} </p><p>When the underlying field of <i>V</i> does not have characteristic 2, then this definition is equivalent to: V ∧ V := V ⊗ V / { v 1 ⊗ v 2 + v 2 ⊗ v 1 ∣ ( v 1 , v 2 ) ∈ V 2 } . {\displaystyle V\wedge V:=V\otimes V{\big /}{\bigl \{}v_{1}\otimes v_{2}+v_{2}\otimes v_{1}\mid (v_{1},v_{2})\in V^{2}{\bigr \}}.} </p><p>The image of v 1 ⊗ v 2 {\displaystyle v_{1}\otimes v_{2}} in the exterior product is usually denoted v 1 ∧ v 2 {\displaystyle v_{1}\wedge v_{2}} and satisfies, by construction, v 1 ∧ v 2 = − v 2 ∧ v 1 {\displaystyle v_{1}\wedge v_{2}=-v_{2}\wedge v_{1}} . Similar constructions are possible for V ⊗ ⋯ ⊗ V {\displaystyle V\otimes \dots \otimes V} (<i>n</i> factors), giving rise to Λ n V {\displaystyle \Lambda ^{n}V} , the <i>n</i>th <a href="/facts/Exterior_power/rdN4TvpR">exterior power</a> of <i>V</i>. The latter notion is the basis of <a href="/facts/Differential_form/T9gyHtMv">differential <i>n</i>-forms</a>. </p><p>The symmetric algebra is constructed in a similar manner, from the <a href="/facts/Symmetric_tensor/CWKBfgnv">symmetric product</a>: V ⊙ V := V ⊗ V / { v 1 ⊗ v 2 − v 2 ⊗ v 1 ∣ ( v 1 , v 2 ) ∈ V 2 } . {\displaystyle V\odot V:=V\otimes V{\big /}{\bigl \{}v_{1}\otimes v_{2}-v_{2}\otimes v_{1}\mid (v_{1},v_{2})\in V^{2}{\bigr \}}.} </p><p>More generally: Sym n V := V ⊗ ⋯ ⊗ V ⏟ n / ( ⋯ ⊗ v i ⊗ v i + 1 ⊗ ⋯ − ⋯ ⊗ v i + 1 ⊗ v i ⊗ … ) {\displaystyle \operatorname {Sym} ^{n}V:=\underbrace {V\otimes \dots \otimes V} _{n}{\big /}(\dots \otimes v_{i}\otimes v_{i+1}\otimes \dots -\dots \otimes v_{i+1}\otimes v_{i}\otimes \dots )} </p><p>That is, in the symmetric algebra two adjacent vectors (and therefore all of them) can be interchanged. The resulting objects are called <a href="/facts/Symmetric_tensor/CWKBfgnv">symmetric tensors</a>. </p> <h2 id="tensor-product-in-programming">Tensor product in programming</h2> <h3>Array programming languages</h3> <p><a href="/facts/Array_programming_languages/gFx1SxIY">Array programming languages</a> may have this pattern built in. For example, in <a href="/facts/APL_programming_language/E2sb98gC">APL</a> the tensor product is expressed as ○.× (for example A ○.× B or A ○.× B ○.× C). In <a href="/facts/J_programming_language/EdIVMBj7">J</a> the tensor product is the dyadic form of */ (for example a */ b or a */ b */ c). </p><p>J's treatment also allows the representation of some tensor fields, as a and b may be functions instead of constants. This product of two functions is a derived function, and if a and b are <a href="/facts/Differentiable_function/jQqTgLk1">differentiable</a>, then a */ b is differentiable. </p><p>However, these kinds of notation are not universally present in array languages. Other array languages may require explicit treatment of indices (for example, <a href="/facts/MATLAB/qPjLISCk">MATLAB</a>), and/or may not support <a href="/facts/Higher-order_function/VjnjGbpt">higher-order functions</a> such as the <a href="/facts/Jacobian_matrix_and_determinant/0tNv6o8j">Jacobian derivative</a> (for example, <a href="/facts/Fortran/m1KqjcMU">Fortran</a>/APL). </p> <h2 id="see-also">See also</h2> Look up <i>tensor product</i> in Wiktionary, the free dictionary. <ul><li><a href="/facts/Dyadics/FoLJnrwR">Dyadics</a> – Second order tensor in vector algebra</li> <li><a href="/facts/Extension_of_scalars/RYGsTsYk">Extension of scalars</a></li> <li><a href="/facts/Monoidal_category/F2l5oKMD">Monoidal category</a> – Category admitting tensor products</li> <li><a href="/facts/Tensor_algebra/H7CKKryf">Tensor algebra</a> – Universal construction in multilinear algebra</li> <li><a href="/facts/Tensor_contraction/fqz0zvWp">Tensor contraction</a> – Operation in mathematics and physics</li> <li><a href="/facts/Topological_tensor_product/eDICDJ2Y">Topological tensor product</a> – Tensor product constructions for topological vector spaces</li></ul> <h2 id="notes">Notes</h2> <ul><li><a href="/facts/Nicolas_Bourbaki/qRJhs9nr">Bourbaki, Nicolas</a> (1989). <i>Elements of mathematics, Algebra I</i>. Springer-Verlag. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 3-540-64243-9.</li> <li><a href="/facts/Tim_Gowers/El3xoEmE">Gowers, Timothy</a>. <a href="https://www.dpmms.cam.ac.uk/~wtg10/tensors3.html">"How to lose your fear of tensor products"</a>. <a href="https://web.archive.org/web/20210507014709/https://www.dpmms.cam.ac.uk/~wtg10/tensors3.html">Archived</a> from the original on 7 May 2021.</li> <li>Grillet, Pierre A. (2007). <i>Abstract Algebra</i>. Springer Science+Business Media, LLC. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 978-0387715674.</li> <li><a href="/facts/Paul_Halmos/mR11lcap">Halmos, Paul</a> (1974). <i>Finite dimensional vector spaces</i>. Springer. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 0-387-90093-4.</li> <li><a href="/facts/Thomas_W._Hungerford/sWS5Wkkw">Hungerford, Thomas W.</a> (2003). <i>Algebra</i>. Springer. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 0387905189.</li> <li><a href="/facts/Serge_Lang/I7MW2rUu">Lang, Serge</a> (2002), <i>Algebra</i>, <a href="/facts/Graduate_Texts_in_Mathematics/VTIfY29j">Graduate Texts in Mathematics</a>, vol. 211 (Revised third ed.), New York: Springer-Verlag, <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 978-0-387-95385-4, <a href="/facts/MR_(identifier)/uP137L11">MR</a> <a href="https://mathscinet.ams.org/mathscinet-getitem?mr=1878556">1878556</a>, <a href="/facts/Zbl_(identifier)/P6rFxKKx">Zbl</a> <a href="https://zbmath.org/?format=complete&q=an:0984.00001">0984.00001</a></li> <li><a href="/facts/Saunders_Mac_Lane/rjgUqfkA">Mac Lane, S.</a>; <a href="/facts/Garrett_Birkhoff/5jeI7Wmr">Birkhoff, G.</a> (1999). <i>Algebra</i>. AMS Chelsea. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 0-8218-1646-2.</li> <li>Aguiar, M.; Mahajan, S. (2010). <i>Monoidal functors, species and Hopf algebras</i>. CRM Monograph Series Vol 29. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 978-0-8218-4776-3.</li> <li><a href="/facts/Fran%25C3%25A7ois_Tr%25C3%25A8ves/cpjQbAov">Trèves, François</a> (2006) [1967]. <i>Topological Vector Spaces, Distributions and Kernels</i>. Mineola, N.Y.: Dover Publications. <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 978-0-486-45352-1. <a href="/facts/OCLC_(identifier)/Yqad8waq">OCLC</a> <a href="https://search.worldcat.org/oclc/853623322">853623322</a>.</li> <li><a href="http://pages.bangor.ac.uk/~mas010/nonabtens.html">"Bibliography on the nonabelian tensor product of groups"</a>.</li></ul> <h2 id="references">References</h2> <ol> <li id="fn:1"><p>Trèves 2006, pp. 403–404. - Trèves, François (2006) [1967]. Topological Vector Spaces, Distributions and Kernels. Mineola, N.Y.: Dover Publications. ISBN 978-0-486-45352-1. OCLC 853623322. <a href="https://search.worldcat.org/oclc/853623322" target="_blank">https://search.worldcat.org/oclc/853623322</a> <a href="#fnref:1" class="footnote-back-ref">↩</a></p></li> <li id="fn:2"><p>Trèves 2006, pp. 403–404. - Trèves, François (2006) [1967]. Topological Vector Spaces, Distributions and Kernels. Mineola, N.Y.: Dover Publications. ISBN 978-0-486-45352-1. OCLC 853623322. <a href="https://search.worldcat.org/oclc/853623322" target="_blank">https://search.worldcat.org/oclc/853623322</a> <a href="#fnref:2" class="footnote-back-ref">↩</a></p></li> <li id="fn:3"><p>Trèves 2006, pp. 407. - Trèves, François (2006) [1967]. Topological Vector Spaces, Distributions and Kernels. Mineola, N.Y.: Dover Publications. ISBN 978-0-486-45352-1. OCLC 853623322. <a href="https://search.worldcat.org/oclc/853623322" target="_blank">https://search.worldcat.org/oclc/853623322</a> <a href="#fnref:3" class="footnote-back-ref">↩</a></p></li> <li id="fn:4"><p>Trèves 2006, pp. 407. - Trèves, François (2006) [1967]. Topological Vector Spaces, Distributions and Kernels. Mineola, N.Y.: Dover Publications. ISBN 978-0-486-45352-1. OCLC 853623322. <a href="https://search.worldcat.org/oclc/853623322" target="_blank">https://search.worldcat.org/oclc/853623322</a> <a href="#fnref:4" class="footnote-back-ref">↩</a></p></li> <li id="fn:5"><p>Hazewinkel, Michiel; Gubareni, Nadezhda Mikhaĭlovna; Gubareni, Nadiya; Kirichenko, Vladimir V. (2004). Algebras, rings and modules. Springer. p. 100. ISBN 978-1-4020-2690-4. <a href="978-1-4020-2690-4" target="_blank">978-1-4020-2690-4</a> <a href="#fnref:5" class="footnote-back-ref">↩</a></p></li> <li id="fn:6"><p>Bourbaki (1989), p. 244 defines the usage "tensor product of x and y", elements of the respective modules. - Bourbaki, Nicolas (1989). Elements of mathematics, Algebra I. Springer-Verlag. ISBN 3-540-64243-9. <a href="#fnref:6" class="footnote-back-ref">↩</a></p></li> <li id="fn:7"><p>Analogous formulas also hold for contravariant tensors, as well as tensors of mixed variance. Although in many cases such as when there is an inner product defined, the distinction is irrelevant. <a href="/wiki/Covariance_and_contravariance_of_vectors" target="_blank">/wiki/Covariance_and_contravariance_of_vectors</a> <a href="#fnref:7" class="footnote-back-ref">↩</a></p></li> <li id="fn:8"><p>"The Coevaluation on Vector Spaces". The Unapologetic Mathematician. 2008-11-13. Archived from the original on 2017-02-02. Retrieved 2017-01-26. <a href="https://unapologetic.wordpress.com/2008/11/13/the-coevaluation-on-vector-spaces/" target="_blank">https://unapologetic.wordpress.com/2008/11/13/the-coevaluation-on-vector-spaces/</a> <a href="#fnref:8" class="footnote-back-ref">↩</a></p></li> <li id="fn:9"><p>See Compact closed category. <a href="/wiki/Compact_closed_category" target="_blank">/wiki/Compact_closed_category</a> <a href="#fnref:9" class="footnote-back-ref">↩</a></p></li> <li id="fn:10"><p>Hungerford, Thomas W. (1974). Algebra. Springer. ISBN 0-387-90518-9. <a href="0-387-90518-9" target="_blank">0-387-90518-9</a> <a href="#fnref:10" class="footnote-back-ref">↩</a></p></li> <li id="fn:11"><p>Chen, Jungkai Alfred (Spring 2004), "Tensor product" (PDF), Advanced Algebra II (lecture notes), National Taiwan University, archived (PDF) from the original on 2016-03-04{{citation}}: CS1 maint: location missing publisher (link) <a href="http://www.math.ntu.edu.tw/~jkchen/S04AA/S04AAL10.pdf" target="_blank">http://www.math.ntu.edu.tw/~jkchen/S04AA/S04AAL10.pdf</a> <a href="#fnref:11" class="footnote-back-ref">↩</a></p></li> <li id="fn:12"><p>Abo, H.; Seigal, A.; Sturmfels, B. (2015). "Eigenconfigurations of Tensors". arXiv:1505.05729 [math.AG]. <a href="/wiki/Anna_Seigal" target="_blank">/wiki/Anna_Seigal</a> <a href="#fnref:12" class="footnote-back-ref">↩</a></p></li> <li id="fn:13"><p>Garrett, Paul (July 22, 2010). "Non-existence of tensor products of Hilbert spaces" (PDF). <a href="https://www-users.cse.umn.edu/~garrett/m/v/nonexistence_tensors.pdf" target="_blank">https://www-users.cse.umn.edu/~garrett/m/v/nonexistence_tensors.pdf</a> <a href="#fnref:13" class="footnote-back-ref">↩</a></p></li> <li id="fn:14"><p>Kadison, Richard V.; Ringrose, John R. (1997). Fundamentals of the theory of operator algebras. Graduate Studies in Mathematics. Vol. I. Providence, R.I.: American Mathematical Society. Thm. 2.6.4. ISBN 978-0-8218-0819-1. MR 1468229. <a href="978-0-8218-0819-1" target="_blank">978-0-8218-0819-1</a> <a href="#fnref:14" class="footnote-back-ref">↩</a></p></li> <li id="fn:15"><p>Tu, L. W. (2010). An Introduction to Manifolds. Universitext. Springer. p. 25. ISBN 978-1-4419-7399-3. <a href="978-1-4419-7399-3" target="_blank">978-1-4419-7399-3</a> <a href="#fnref:15" class="footnote-back-ref">↩</a></p></li> </ol>