Is tensor product the same as dyadic product of two vectors?

In summary, the tensor product is a way of combining two vector spaces into a new vector space, where bilinear maps become linear. It can be thought of as a multilinear function and is used in physics to encode multiple pieces of information into one object. The tensor product can also be viewed as a bilinear map itself, and is closely related to linear transformations in dual spaces. It is a category-theoretical construction and has physical intuition when considering rank-2 dyadic tensors.
  • #1
xopek
24
0
Is tensor product the same as dyadic product of two vectors? And dyadic multiplication is just matrix multiplication? You have a column vector on the left and a row vector on the right and you just multiply them and that's it? We just create a matrix out of two vectors so we encode two different things, such a stress tensor in different directions? Sounds too simple to be true. I tried reading about the tensor product space and it was way too abstract and every source gives a different explanation.
 
Physics news on Phys.org
  • #2
xopek said:
Is tensor product the same as dyadic product of two vectors?
The dyadic product is a tensor product, but a tensor product can have more than two components and is in general also the sum of e.g. dyadic products.

xopek said:
And dyadic multiplication is just matrix multiplication?
Yes. Column times row.
xopek said:
You have a column vector on the left and a row vector on the right and you just multiply them and that's it? We just create a matrix out of two vectors so we encode two different things, such a stress tensor in different directions? Sounds too simple to be true. I tried reading about the tensor product space and it was way too abstract and every source gives a different explanation.
Try this one:
https://www.physicsforums.com/insights/what-is-a-tensor/

What is a derivative? There are plenty of ways to consider it: slope, linear function, limit, etc. And like
$$
D_2(x^2)=\left.\dfrac{d}{dx}\right|_{x=2} (x^2)=4
$$
can be viewed as the linear function ##x\longmapsto 4\cdot x## can tensors be considered as multilinear functions. This is especially in physics the case. So although they are simply a number scheme, they are simultaneously functions, too.
 
  • #3
At the vector space level, the tensor product of two vector spaces over the same field is a new vector space
over which all bilinear maps into a fixed 3rd vector space becomes linear. It's one of those category-theoretical constructions. But maybe you were looking for more Physical intuition?
 
  • #4
WWGD said:
a new vector space over which all bilinear maps into a fixed 3rd vector space becomes linear.
So the tensor product creates a mathematical object called a tensor? And that object is also a bilinear map? And in case of two components it is also a 2-dimensional matrix? I know that typically a 2-dimensional matrix is a linear transformation map in a 2D vector space such as a rotation map that takes a vector and spits out another vector. But in our case this matrix (our dyadic tensor) is a bilinear map which takes two input vectors and spits out a scalar. I know that a bilinear map can become linear with one of the components held constant, i.e. what happens in a dual space V* where a covariant vector v* takes a contravariant vector v from V and produces a scalar. Is this what happens in case of the tensor product space as well?

WWGD said:
But maybe you were looking for more Physical intuition?
Yes, that too. I was thinking of something simple such as rank-2 dyadic tensor, say in two dimensions, when two 2D vectors are "mixed" together into a 2x2 matrix aka dyadic tensor which then encodes two pieces of information coming from the two vectors, say, an angle and a direction. My main confusion is the distinction between some other object acting on a tensor, and a tensor acting on other objects. And pretty much everything about tensors.
 
  • #5
Yes, the tensor product $$ V \otimes W$$ of two vector spaces $$V, W $$ over the same field is a vector space whose dimension is the product of the dimensions of $$V \and W $$, so that every bilinear map $$ B: V \times W \rightarrow Z$$ factors into a linear map $$L$$ from $$ V \otimes W \rightarrow Z $$
So that the diagram commutes. Sorry, I don't know how to implement diagrams here in PF.
 

Related to Is tensor product the same as dyadic product of two vectors?

1. What is a tensor product and dyadic product?

A tensor product is a mathematical operation that combines two vectors or matrices to create a new vector or matrix. A dyadic product is a special case of the tensor product where the two vectors are of the same dimension and result in a square matrix.

2. Are tensor product and dyadic product the same thing?

No, they are not the same thing. While the dyadic product is a type of tensor product, not all tensor products are dyadic products. The main difference is that a dyadic product always results in a square matrix, while a tensor product can result in a matrix of any dimension.

3. How is the tensor product different from the dot product?

The tensor product and dot product are two different mathematical operations. The dot product is a scalar value that results from multiplying two vectors, while the tensor product results in a new vector or matrix.

4. What are the applications of the tensor product and dyadic product?

The tensor product and dyadic product have various applications in mathematics, physics, and engineering. They are used in vector and matrix operations, quantum mechanics, and in the study of tensors and their properties.

5. Is the tensor product commutative?

No, the tensor product is not commutative. This means that the order in which the two vectors are multiplied matters and can result in different outputs. However, the dyadic product is commutative since it always results in a square matrix.

Similar threads

  • Linear and Abstract Algebra
Replies
7
Views
576
  • Linear and Abstract Algebra
Replies
2
Views
1K
  • Linear and Abstract Algebra
Replies
1
Views
883
  • Linear and Abstract Algebra
Replies
33
Views
1K
  • Linear and Abstract Algebra
Replies
1
Views
1K
Replies
5
Views
3K
  • Quantum Physics
Replies
11
Views
1K
  • Linear and Abstract Algebra
Replies
9
Views
546
Replies
4
Views
1K
Replies
14
Views
2K
Back
Top