Tensor products

1. Formula
   Observation

   The slide text states, "Let U, V, and W be linear spaces over the number field P."

U, V, W, P

linear space over the number field P

linear space

1. Formula
   Observation

   The slide text states, "bilinear mapping $\varphi$."

$\varphi$

bilinear mapping from $U \times V$ to W

mapping

1. Formula
   Observation

   The slide text states, "any bilinear mapping f."

f

bilinear mapping from $U \times V$ to any linear space P over P

mapping

1. Formula
   Observation

   The slide text states, "unique linear mapping $\tau$."

$\tau$

unique linear mapping from W to P

mapping

1. Formula
   Observation

   The slide text displays "(W', $\varphi$')".

W', $\varphi$'

Another tensor product linear space and its bilinear map.

Linear spaces and mappings.

1. Formula
   Observation

   The slides display U, V, W and the number field P multiple times; the presenter explains that U, V, and W are linear spaces over the number field P.

U, V, W, $\mathbb{P}$

The linear spaces U, V, W over the number field $\mathbb{P}$, and the scalar field from which these spaces draw their scalars.

U, V, W are linear spaces; $\mathbb{P}$ is the scalar field.

1. Formula
   Observation

   The annotation page states: "We denote by $U \otimes V$ the unique tensor product (W, $\varphi$) of U and V up to isomorphism."

2. Audio
   Observation

   The presenter says: "We denote the unique tensor product of U and V up to isomorphism by U tensor V."

$U \otimes V$

Notation for the tensor product space of U and V uniquely determined up to isomorphism.

Denotes a linear space, not a single tensor element.

1. Formula
   Observation

   The proposition page states: "(W, $\varphi$) and (W', $\varphi$') are two tensor products of U and V."

2. Formula
   Observation

   The annotation page states: "the unique tensor product (W, $\varphi$)."

(W, $\varphi$)

A tensor product structure where W is the underlying space and $\varphi$ is the associated bilinear map.

W is a linear space, $\varphi$: $U \times V \to W$

1. Formula
   Observation

   The proposition page displays "(W', $\varphi$')".

2. Diagram
   Observation

   Label the arrow from $U \times V$ to W' in the commutative diagram as $\varphi$'.

(W', $\varphi$')

Another tensor product structure of U and V.

W' is a vector space, and $\varphi$': $U \times V \to W$'.

1. Formula
   Observation

   $\varphi(u,v)$ appears on both the proposition page and the annotation page.

2. Audio
   Observation

   The presenter says, "This $\varphi$ applied to u and v can be written directly as u tensor v."

$\varphi$

The bilinear map in the tensor product structure that sends (u,v) to the element in the tensor product.

The map $\varphi$ from the Cartesian product of U and V to W.

1. Formula
   Observation

   The proposition page states, "There exists a unique vector space isomorphism $\tau$ from W to W'."

2. Diagram
   Observation

   In the commutative diagram, the dashed arrow from W to W' is labeled $\tau$.

$\tau$

The unique vector space isomorphism connecting the two tensor product spaces.

The map $\tau$ from W to W'.

1. Formula
   Observation

   The annotation page states, "For any $u \in U$ and $v \in V$, $\varphi(u,v)$ is denoted $u \otimes v$."

2. Audio
   Observation

   The instructor says, "For any u in U and v in V, applying $\varphi$ to u and v is simply written as u tensor v."

$u \otimes v$

A basic tensor obtained from $u \in U$ and $v \in V$ via the tensor product map.

u is in U and v is in V

1. Formula
   Observation

   The slide shows the definition of tensor product and its commutative diagram.

2. Audio
   Observation

   The instructor verbally states the definition of tensor product and its universal property.

Let U, V, W be linear spaces over the field P. If there exists a bilinear map $\varphi$ from $U \times V$ to W such that for any bilinear map f from $U \times V$ to an arbitrary linear space P over P, there exists a unique linear map $\tau$ from W to P making the diagram commute, i.e., $f = \tau \varphi$, then the pair (W, $\varphi$) is called a tensor product of U and V.

1. U, V, W are linear spaces over the field P.

2. $\varphi$ is a bilinear map from $U \times V$ to W.

3. f is a bilinear map from $U \times V$ to an arbitrary linear space P over P.

4. $\tau$ is the unique linear map from W to P.

1. Audio
   Observation

   The instructor mentions, "This diagram—and this property—has a special name in algebra: it is called the universal property, or sometimes the universal property."

The commutative diagram property described in the definition of the tensor product is called a universal property in algebra. It states that although a tensor product satisfying this property is not unique, it is unique up to isomorphism.

1. Definition of tensor product.

1. Formula
   Observation

   Annotation page: "We denote the unique tensor product of U and V up to isomorphism by $U \otimes V$, written as (W, $\varphi$)."

2. Audio
   Observation

   The presenter says, "The tensor product is unique up to isomorphism," and then adds, "We denote this unique tensor product of U and V up to isomorphism by $U \otimes V$."

Since the preceding proposition states that there exists a unique isomorphism between any two tensor product structures, we may regard the tensor product of U and V as a unique object up to isomorphism, denoted by the symbol $U \otimes V$.

1. U and V are linear spaces over the number field $\mathbb{P}$

2. Here $U \otimes V$ refers to the tensor product space uniquely determined up to isomorphism

1. Formula
   Observation

   Annotation page: "For any $u \in U$ and $v \in V$, $\varphi(u,v)$ is denoted by $u \otimes v$."

2. Audio
   Observation

   The presenter says, "For any u in U and v in V, the value of $\varphi$ at u and v is simply written as $u \otimes v$."

Given the bilinear map $\varphi$ in a tensor product structure, for any $u \in U$ and $v \in V$, $\varphi(u,v)$ is written as $u \otimes v$. This notation represents the elementary tensor generated by the two vectors.

1. u is in U

2. v belongs to V

3. φ is the bilinear map in the tensor product construction

1. $U \otimes V$ denotes the unique tensor product up to isomorphism

1. Formula
   Observation

   The property proposition lists three items: ($a_1u_1 + a_2u_2$) ⊗ $v = a_1u_1$ ⊗ $v + a_2u_2$ ⊗ v; u ⊗ ($b_1v_1 + b_2v_2$) = $b_1u$ ⊗ $v_1 + b_2u$ ⊗ $v_2$; a(u ⊗ v) = (au) ⊗ $v = u$ ⊗ (av).

2. Audio
   Observation

   The instructor says, 'Because we know the tensor product is constructed via a bilinear function, the tensor product inherits bilinearity, and thus has the following properties.'

The tensor product notation satisfies distributivity over addition and scalar multiplication in both the left and right arguments, and scalars may be moved from one side to the other. These properties follow directly from the fact that the tensor product is constructed from a bilinear map.

1. u, u₁, and u₂ all belong to U

2. v, v₁, and v₂ all belong to V

3. a, $a_1$, $a_2$, $b_1$, $b_2$ are scalars in the field ℙ

1. elementary tensor notation $u \otimes v$

1. Formula
   Observation

   Example page: 'Let A be an $m \times n$ matrix and $B a p \times q$ matrix. Then the Kronecker product of A and B (denoted A ⊗ B) is a tensor product: it is an mp × nq block matrix.' A block matrix expression is provided.

2. Audio
   Observation

   The instructor says, 'Then what about the Kronecker product of A and B? It is denoted A tensor B, and it is also a tensor product — specifically, an mp by nq block matrix.'

When A is an m by n matrix and B is a p by q matrix, the Kronecker product of A and B is denoted by $A \otimes B$; it is an mp by nq block matrix, each block being the product of an element of A with the entire matrix B.

1. A is an m by n matrix

2. B is a p by q matrix

3. a_{ij} is an element of A

1. elementary tensor notation $u \otimes v$

1. Formula
   Observation

   Problem page: "Let U, V, W be linear spaces over the number field P. Then $\hom(U, \hom(V, W)) \cong \hom(U \otimes V, W)$."

2. Audio
   Observation

   After explaining the meaning of Hom spaces, the instructor says, "It is isomorphic to the linear space consisting of all linear maps from the tensor product space $U \otimes V$ to W," and calls this "a duality relation between Hom and tensor product."

For three linear spaces U, V, W over a number field, the linear space of linear maps from U to the linear space of linear maps from V to W is isomorphic to the linear space of linear maps from the tensor product space $U \otimes V$ to W. This reflects an adjoint/duality relation between Hom and tensor product.

1. U, V, W are linear spaces over the number field $\mathbb{P}$

1. $U \otimes V$ denotes the unique tensor product up to isomorphism

1. Formula
   Observation

   The slide displays the proposition text and its commutative diagram

2. Audio
   Observation

   The instructor verbally states the proposition

Let U and V be linear spaces over the number field P. If (W, $\varphi$) and (W', $\varphi$') are two tensor products of U and V, then there exists a unique linear space isomorphism $\tau$ from W to W' such that the following diagram commutes. That is, W' = $\tau(W)$ and $\varphi$' = $\tau \varphi$, so (W', $\varphi$') = ($\tau(W)$, $\tau \varphi$).

1. U and V are linear spaces over the number field P

2. (W, $\varphi$) and (W', $\varphi$') are two tensor products of U and V

there exists a unique

1. Formula
   Observation

   Proposition page: "Let U, V be linear spaces over the number field P. If (W,$\varphi$) and (W',$\varphi$') are two tensor products of U and V, then there exists a unique linear space isomorphism $\tau$ from W to W' making the diagram commute. That is, W'=$\tau(W)$ and $\varphi$'=$\tau\varphi$, so (W',$\varphi$')=($\tau(W)$,$\tau\varphi$)."

2. Diagram
   Observation

   The commutative diagram shows $U \times V$ mapping to W via $\varphi$ and to W' via $\varphi$', with a dashed isomorphism $\tau$ from W to W'.

3. Audio
   Observation

   The instructor says, "This means that the tensor product is unique up to isomorphism."

If (W,$\varphi$) and (W',$\varphi$') are both tensor products of U and V, then there exists a unique linear space isomorphism $\tau$:W\to W' making the diagram commute, i.e., W'=$\tau(W)$ and $\varphi$'=$\tau\varphi$.

1. U and V are linear spaces over the number field $\mathbb{P}$.

2. (W,$\varphi$) is a tensor product of U and V.

3. (W',$\varphi$') is another tensor product of U and V.

For any two tensor product structures satisfying the premises, there exists a unique isomorphism $\tau$.

1. Audio
   Observation

   The presenter says, "Because we know the tensor product is derived from a bilinear map, the tensor product inherits bilinearity, and thus has the following properties."

2. Formula
   Observation

   The slide directly lists three bilinear properties.

1. Expression

   $$$$
   Explanation

   First point out that the tensor product is constructed from a bilinear function.

   Justification

   The presenter verbally states, "The tensor product is derived from a bilinear function."

   Shown in the video
2. Expression

   $$(a_1u_1+a_2u_2)\otimes v=a_1u_1\otimes v+a_2u_2\otimes v$$
   Explanation

   This yields linearity in the first variable.

   Justification

   A bilinear map is linear in each variable separately.

   Shown in the video
3. Expression

   $$u\otimes(b_1v_1+b_2v_2)=b_1u\otimes v_1+b_2u\otimes v_2$$
   Explanation

   Then obtain linearity in the second variable.

   Justification

   A bilinear map is linear in each variable separately.

   Shown in the video
4. Expression

   $$a(u\otimes v)=(au)\otimes v=u\otimes(av)$$
   Explanation

   A scalar can act on the entire tensor, or it can be moved onto either vector.

   Justification

   This is the scalar compatibility of bilinearity expressed in tensor notation.

   Shown in the video

The tensor product notation satisfies the three listed bilinear properties.

1. Audio
   Observation

   The instructor explains step by step: 'Hom(U, V) denotes the set of all linear maps from U to V... which itself forms a vector space... then taking Hom again... it is isomorphic to the vector space of all linear maps from the tensor product of U and V to W.'

2. Formula
   Observation

   The problem page gives $\hom(U, \hom(V, W)) \cong \hom(U \otimes V, W)$.

1. The video does not provide the explicit construction or proof steps for this isomorphism; interested viewers are invited to derive it themselves.

1. Expression

   $$$$
   Explanation

   First interpret $\hom(V,W)$ as 'the set of all linear maps from V to W' and regard it as a vector space.

   Justification

   The instructor verbally defines the Hom space.

   Shown in the video
2. Expression

   $$$$
   Explanation

   Next, interpret $\hom(U,\hom(V,W))$ as the linear space of all linear maps from U to this linear space.

   Justification

   The presenter continues explaining the meaning of nested Hom.

   Shown in the video
3. Expression

   $$\hom(U, \hom(V, W)) \cong \hom(U \otimes V, W)$$
   Explanation

   Finally, it is stated that this space is isomorphic to the space of all linear maps from $U \otimes V$ to W.

   Justification

   The formula on the problem slide matches the presenter’s verbal explanation.

   Shown in the video

This equation expresses the adjoint/duality relationship between Hom and tensor product, but the video does not show the proof.

1. Formula
   Observation

   The example slide gives the dimensions of A and B, and the block matrix form of $A \otimes B$.

2. Audio
   Observation

   The presenter recalls 'that earlier example' and repeats that $A \otimes B$ is an mp $\times$ nq block matrix.

Let A be an $m \times n$ matrix and $B a p \times q$ matrix; write the Kronecker product of A and B and state its size.

1. A is an $m \times n$ matrix

2. B is $a p \times q$ matrix

3. a_{ij} denotes the entry of A

Give the expression for A tensor B and its matrix dimensions.

1. Expression

   $$A \otimes B=\begin{pmatrix} a_{11}B & \cdots & a_{1n}B \\ \vdots & & \vdots \\ a_{m1}B & \cdots & a_{mn}B \end{pmatrix}$$
   Explanation

   Replace each element a ij of A with the block a ij times B.

   Justification

   The example page directly gives the definition of the Kronecker product.

   Shown in the video
2. Expression

   $$mp \times nq$$
   Explanation

   The resulting block matrix has 〔m〕 row blocks and 〔n〕 column blocks, each of size 〔p〕 $\times$ 〔q〕, so the total size is 〔mp〕 $\times$ 〔nq〕.

   Justification

   Both the presenter and the slide state that it is an 〔mp〕 $\times$ 〔nq〕 block matrix.

   Shown in the video

A tensor B is the above block matrix and is an m p by n q matrix.

The total size can be verified by multiplying the number of blocks 〔m〕 $\times$ 〔n〕 by the size of each block 〔p〕 $\times$ 〔q〕.

1. Formula
   Observation

   The problem page states: Let U, V, W be linear spaces over the number field P, then hom of U into hom of V into W is isomorphic to hom of U tensor V into W.

2. Audio
   Observation

   The presenter explains the meaning of this formula and says, "If you are interested, you may try to prove it yourself."

1. The video does not provide a proof nor construct an explicit mapping.

Let U, V, W be linear spaces over the number field P. Explain the relationship between hom of U into hom of V into W and hom of U tensor V into W.

1. U, V, and W are linear spaces over the number field $\mathbb{P}$.

Determine whether two Hom spaces are isomorphic and understand the meaning of this isomorphism.

1. Expression

   $$\hom(V,W)$$
   Explanation

   First interpret it as the linear space consisting of all linear maps from V to W.

   Justification

   The instructor gives an oral definition of the Hom space.

   Shown in the video
2. Expression

   $$\hom(U,\hom(V,W))$$
   Explanation

   Then interpret it as the space consisting of all linear maps from U to the above linear space.

   Justification

   The instructor continues explaining nested Hom.

   Shown in the video
3. Expression

   $$\hom(U \otimes V,W)$$
   Explanation

   The problem's conclusion states that it is isomorphic to the linear space of linear maps from the tensor product space $U \otimes V$ to W.

   Justification

   The formula on the problem slide directly gives the isomorphism relation.

   Shown in the video

The linear space of linear maps from U to V to W is isomorphic to the linear space of linear maps from the tensor product of U and V to W.

The video does not demonstrate verification; it only suggests that viewers may prove it themselves.

1. Diagram
   Observation

   The slide displays the mapping relationships among $U \times V$, W, and P

1. U times V

2. W

3. P

4. $\varphi$

5. f

6. $\tau$

1. The diagram structure remains unchanged

Solid arrows represent the given bilinear maps $\varphi$ and f; dashed arrows represent the existence and uniqueness of the linear map $\tau$. The diagram visually illustrates the composition relation $f = \tau \varphi$.

1. Diagram
   Observation

   The slide shows a mapping diagram among $U \times V$, W, and W'

1. U times V

2. W

3. W prime

4. phi

5. phi prime

6. tau

1. The diagram structure remains unchanged.

Solid arrows represent the bilinear maps phi and phi prime between two tensor products; dashed arrows represent the existence and uniqueness of the isomorphism tau. The diagram visually illustrates the relation phi prime equals tau composed with phi.

1. Diagram
   Observation

   The screen displays a commutative triangle diagram: $U \times V$ maps to W via $\varphi$, maps to W' via $\varphi$', and there is a dashed arrow $\tau$ from W to W'.

2. Formula
   Observation

   The proposition states in words: 'so that the diagram commutes. That is, W' = $\tau(W)$ and $\varphi$' = $\tau\varphi$.'

1. Cartesian product of U and V

2. W

3. W'

4. $\varphi$

5. $\varphi$'

6. $\tau$

1. The screen freezes, showing two maps from $U \times V$ to W and to W'

2. A dashed isomorphism arrow from W to W' is shown simultaneously

1. The relative positions of the three objects $U \times V$, W, and W' remain unchanged

2. The commutativity of the diagram is fixed by the statement of the proposition

This diagram intuitively expresses that any two tensor product structures can be connected by a unique isomorphism $\tau$ such that composing $\varphi$ with $\tau$ yields the same result as $\varphi$' directly.

1. Animation
   Observation

   The slide transitions sequentially from 'Proposition + Commutative Diagram' to 'Annotation', then adds the 'Proposition Property Box', next switches to 'Example', and finally adds the 'Exercise Box'.

2. Caption evidence
   Observation

   At the bottom of the page, the date '2022 year 3 month 2 day' is visible, and the page number changes from $3/5$ to $4/5$, then to $5/5$.

1. Proposition slide

2. Annotation slide

3. Property proposition box

4. Example slide

5. Exercise box

1. At approximately 00:08, transition from the uniqueness proposition slide to the annotation slide.

2. At approximately 00:21, add the bilinearity property proposition box below the annotation slide.

3. At approximately 00:40, switch to the Kronecker product example slide.

4. At approximately 00:56, add the Hom isomorphism exercise box below the example slide.

1. The presentation is a static slideshow with no additional animations demonstrating the computation steps.

2. The layout consistently retains the heading 'Background Knowledge'.

Visual progression follows the lecture order: first stating the uniqueness of isomorphism, then introducing notation and basic properties, followed by a matrix example, and finally supplementing the relationship between Hom and tensor product.

1. Audio
   Observation

   The instructor mentions, "Generally speaking, the postgraduate entrance examination does not test the concept of tensor products... However, you may still apply some related knowledge of tensor products to solve problems."

Tensor products are considered outside the scope of linear algebra.

Tensor products belong within the scope of linear algebra; although they are rarely tested directly in foundational examinations, related knowledge can be used to solve specific problems.

1. Formula
   Observation

   The annotation page simultaneously displays '$U \otimes V$ denotes the unique tensor product (W,$\varphi$)' and '$\varphi(u,v)$ is denoted $u \otimes v$'.

2. Audio
   Observation

   The presenter uses 'U tensor V' to refer to the space and 'u tensor v' to refer to the image under the mapping.

Seeing the same $\otimes$ symbol, one might mistakenly assume that $U \otimes V$ and $u \otimes v$ are objects of the same kind.

$U \otimes V$ denotes the entire tensor product space; $u \otimes v$ denotes a basic tensor element in that space generated by u and v.

1. Audio
   Observation

   The presenter emphasizes that 'Hom U V means the set of all linear maps from U to V' and adds that 'it can also form a vector space'.

One might mistakenly interpret $\hom(U,V)$ as a specific linear map.

$\hom(U,V)$ is the set of all linear maps from U to V, and itself forms a vector space.

1. Audio
   Observation

   The instructor refers to the diagrammatic property in the definition as a universal property.

The universal property is the defining characteristic of the tensor product.

1. Audio
   Observation

   The instructor derives a proposition from the definition.

The uniqueness-up-to-isomorphism proposition for tensor products relies on the universal-property definition of tensor products.

1. Formula
   Observation

   The proposition slide first states the uniqueness-up-to-isomorphism result; the annotation slide then defines 'let $U \otimes V$ denote … the unique tensor product'.

2. Audio
   Observation

   The instructor first says 'the tensor product is unique up to isomorphism', then says 'so we use U tensor V to denote …'.

Only because any two tensor product constructions are unique up to isomorphism can we adopt $U \otimes V$ as the standard notation up to isomorphism.

1. Formula
   Observation

   On the same annotation slide, the space notation $U \otimes V$ is defined first, then the element notation $u \otimes v$ is defined.

After fixing the tensor product space notation, the notation for its elementary elements is further defined.

1. Audio
   Observation

   After defining $u \otimes v$, the instructor says 'since we know the tensor product arises from a bilinear map, the tensor product inherits bilinearity'.

2. Formula
   Observation

   The slide then lists three properties of $\otimes$.

Once the elementary tensor notation is introduced, the properties of bilinear maps are rewritten as operational rules for $\otimes$.

1. Animation
   Observation

   The page shifts from bilinearity to the Kronecker product example.

2. Audio
   Observation

   The instructor says 'and let us recall that earlier example'.

1. The video does not explicitly verify the bilinearity properties step by step for the Kronecker product; it only revisits it as an example of the tensor product.

The matrix Kronecker product serves as a concrete instance of the tensor product concept.

1. Formula
   Observation

   The title page directly displays $\hom(U, \hom(V, W)) \cong \hom(U \otimes V, W)$.

2. Audio
   Observation

   The instructor calls this 'a duality relation between Hom and tensor.'

The tensor product space notation $U \otimes V$ is used to express the isomorphism relation between Hom and the tensor product.

1. Audio
   Observation

   The instructor says, 'This duality relation includes, in homological algebra later on, a similar concept involving adjoint functors—this is an adjoint isomorphism.'

1. The video only mentions subsequent content orally, without elaborating formal definitions of homological algebra or adjoint functors.

The instructor broadly connects this isomorphism to the idea of adjoint functors or adjoint isomorphisms that will be studied later.

1. Formula
   Observation

   The slide displays the definition of tensor products.

1. Definition of tensor product.
2. universal property

1. Formula
   Observation

   The slide displays the proposition on uniqueness-up-to-isomorphism of tensor products.

1. isomorphism uniqueness of the tensor product

1. Formula
   Observation

   The proposition page states that there exists a unique isomorphism between two tensor products making the diagram commute.

2. Audio
   Observation

   The instructor concludes, 'The tensor product is unique up to isomorphism.'

1. The tensor product is unique up to isomorphism.
2. $U \otimes V$ denotes the unique tensor product up to isomorphism

1. Formula
   Observation

   The annotation page defines $U \otimes V$ as the unique tensor product up to isomorphism.

1. $U \otimes V$ denotes the unique tensor product up to isomorphism
2. elementary tensor notation $u \otimes v$

1. Audio
   Observation

   The instructor says the tensor product is derived from bilinear maps, so it possesses bilinearity.

2. Formula
   Observation

   The slide lists three properties.

1. Bilinearity of the tensor product
2. Deriving tensor product properties from bilinear maps.

1. Formula
   Observation

   The example page states that A tensor B is an m p by n q block matrix.

1. Kronecker product of matrices as an example of tensor products
2. Kronecker product example

1. Formula
   Observation

   The problem page gives 〔$\hom(U, \hom(V, W))$〕 $\cong$ 〔$\hom(U \otimes V, W)$〕.

2. Audio
   Observation

   The presenter explains this as a duality relationship between Hom and tensor product.

1. Isomorphism relation between Hom and tensor product
2. Explain the meaning of the isomorphism between Hom and the tensor product.
3. Hom and tensor product isomorphism problem