Natural Isomorphism b/w Dual Spaces Tensor Prod & Multilinear Form Space

Click For Summary

Discussion Overview

The discussion revolves around proving a natural isomorphism between the tensor product of dual spaces of finite-dimensional vector spaces and the space of multilinear forms. Participants explore methods to demonstrate the injectivity or surjectivity of a specific linear map, with a focus on the implications of dimensionality and the necessity of choosing bases.

Discussion Character

  • Technical explanation, Debate/contested, Mathematical reasoning

Main Points Raised

  • Some participants propose defining a map \( A \) from \( V_1^*\times\cdots\times V_k^* \) to \( \mathcal L^k(V_1, \ldots, V_k; F) \) and using its properties to establish the isomorphism.
  • Others argue that showing the kernel of the linear map \( \tilde A \) is trivial is necessary for proving the isomorphism, but express a desire to do so in a basis-free manner.
  • A later reply suggests that demonstrating the surjectivity of \( \tilde A \) might be a more straightforward approach, relating it to the representation of matrices as sums of rank 1 matrices.
  • One participant expresses doubt about achieving the proof without selecting a basis, indicating a belief that the result may not hold in infinite-dimensional spaces.
  • Another participant concurs, emphasizing the need to utilize the finite-dimensionality of the spaces involved.

Areas of Agreement / Disagreement

Participants generally agree that the proof may require the selection of bases, especially given the context of finite-dimensional spaces. However, there is no consensus on whether the proof can be completed without this step, and the discussion remains unresolved regarding the necessity of basis selection.

Contextual Notes

The discussion highlights the limitations of the argument in infinite-dimensional spaces and the dependence on the dimensionality of the vector spaces involved. There are unresolved aspects regarding the kernel of the map and the implications of surjectivity versus injectivity.

caffeinemachine
Gold Member
MHB
Messages
799
Reaction score
15
I am trying to prove the following.

Let $V_1, \ldots, V_k$ be finite dimensional vector spaces over a field $F$.
There is a natural isomorphism between $V_1^*\otimes\cdots\otimes V_k^*$ and $\mathcal L^k(V_1, \ldots, V_k;\ F)$.

Define a map $A:V_1^*\times\cdots\times V_k^*\to \mathcal L^k(V_1, \ldots, V_k;\ F)$ as
\begin{equation*}
A(\omega_1, \ldots, \omega_k)(v_1, \ldots, v_k)=\omega(v_1)\cdots\omega_k(v_k)
\end{equation*}
for all $(\omega_1, \ldots, \omega_k)\in V_1^*\times\cdots\times V_k^*$.
It can be seen that $A$ is a multilinear map.
By the universal property of tensor product, there exists a unique linear map $\tilde A: V_1^*\otimes\cdots\otimes V_k^*\to \mathcal L^{k}(V_1, \ldots, V_k; \ F)$ such that $\tilde A\circ \pi=A$.

We also know that
\begin{equation*}
\dim V_1^*\otimes\cdots\otimes V_k^*=\dim \mathcal L^k(V_1, \ldots, V_k; \ F)
\end{equation*}
Thus we just need to show that $\ker \tilde A=0$ to show that $V_1^*\otimes\cdots\otimes V_k^*$ and $\mathcal L^{k}(V_1, \ldots, V_k; \ F)$ are isomorphic.

My Problem: I want to show the triviality of the kernel in a basis free manner. But here I am stuck.

Can anybody help?
 
Physics news on Phys.org
caffeinemachine said:
I am trying to prove the following.

Let $V_1, \ldots, V_k$ be finite dimensional vector spaces over a field $F$.
There is a natural isomorphism between $V_1^*\otimes\cdots\otimes V_k^*$ and $\mathcal L^k(V_1, \ldots, V_k;\ F)$.

Define a map $A:V_1^*\times\cdots\times V_k^*\to \mathcal L^k(V_1, \ldots, V_k;\ F)$ as
\begin{equation*}
A(\omega_1, \ldots, \omega_k)(v_1, \ldots, v_k)=\omega(v_1)\cdots\omega_k(v_k)
\end{equation*}
for all $(\omega_1, \ldots, \omega_k)\in V_1^*\times\cdots\times V_k^*$.
It can be seen that $A$ is a multilinear map.
By the universal property of tensor product, there exists a unique linear map $\tilde A: V_1^*\otimes\cdots\otimes V_k^*\to \mathcal L^{k}(V_1, \ldots, V_k; \ F)$ such that $\tilde A\circ \pi=A$.

We also know that
\begin{equation*}
\dim V_1^*\otimes\cdots\otimes V_k^*=\dim \mathcal L^k(V_1, \ldots, V_k; \ F)
\end{equation*}
Thus we just need to show that $\ker \tilde A=0$ to show that $V_1^*\otimes\cdots\otimes V_k^*$ and $\mathcal L^{k}(V_1, \ldots, V_k; \ F)$ are isomorphic.

My Problem: I want to show the triviality of the kernel in a basis free manner. But here I am stuck.

Can anybody help?
Instead of trying to show that $\tilde A$ is injective, I think it would be easier to show that it is surjective. This should somehow be equivalent to the fact that a $k\times k$ matrix is a sum of rank $1$ matrices.
 
Opalg said:
Instead of trying to show that $\tilde A$ is injective, I think it would be easier to show that it is surjective. This should somehow be equivalent to the fact that a $k\times k$ matrix is a sum of rank $1$ matrices.
Hello Opalg,

Sorry for the late reply. I somehow forgot about this post.

I can show that $A$ is surjective by choosing a basis. I am getting more and more convinced that this cannot be done without choosing a basis.
 
caffeinemachine said:
I am getting more and more convinced that this cannot be done without choosing a basis.
I tend to agree. For one thing, I believe that the result is false if the spaces are infinite-dimensional. So you somehow need to make use of the fact that the spaces are finite-dimensional, and the obvious way is to choose bases for them.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
2
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K