张量习题

1.Prove that $v_1,\dots ,v_r\in V^{\ast }$ is linearly dependent if and only if
$$v_1\wedge \cdots \wedge v_r=0$$

Proof:
If $v^1,\dots .v^r$ is linearly dependent, $\exists v^r=a_1{v}^1+\cdots +a_{r-1}{v}^{r-1}$, thus
$$v^1\wedge \cdots \wedge v^r=\sum _{\lambda =1}^{r-1}{a}_{\lambda }{v}^1\wedge \cdots \wedge v^{r-1}\wedge v^{\lambda }=0$$

If $v^1,\dots ,v^r$ is linearly independent, we can enlarge them to be a basis of V ∗ : { v 1 , … , v r , … , v n } . V^*:\{v^1,\dots,v^r,\dots,v^n\}. V∗:{v1,…,vr,…,vn}. { e i } \{e_i\} {ei​} is the conjugate basis of { v i } \{v^i\} {vi} in $V$, thus
$$\begin{array}{rl}{v}^1\wedge \cdots \wedge v^r(e_1,\dots ,e_r)& ={\delta }_{i_1\dots i_r}^{1\dots r}{v}^{i_1}(e_1,\dots ,e_r)\\ & =\left(\begin{array}{ccc}{v}^1(e_1)& \dots & v^1(e_r)\\ ⋮& & ⋮\\ v^r(e_1)& \dots & v^r(e_r)\end{array}\right)\\ & =1\end{array}$$

thus, $v^1\wedge \cdots \wedge v^r\ne 0.$

2.Let $F:V\to W$ be a linear map, prove that for $\Phi \in A^r(W)$ and $\Psi \in A^s(W)$
$$F^{\ast }(\Phi \wedge \Psi )=F^{\ast }\Phi \wedge F^{\ast }\Psi .$$

Proof:

$\forall u_1,\dots ,u_{r+s}\in V$, we have:
$$\begin{array}{rl}& (F^{\ast }(\Phi \wedge \Psi ))(u_1,\dots ,u_{r+s})\\ & =(\Phi \wedge \Psi )(F(u_1),\dots ,F(u_{r+s}))\\ & =\frac{1}{r!s!}{\delta }_{1\dots r+s}^{i_1\dots i_{r+s}}\Phi (F(u_{i_1}),\dots ,F(u_{i_r}))\cdot \Psi (F(u_{i_{r+1}}),\dots ,F(u_{i_{r+s}}))\\ & =(F^{\ast }\Phi )\wedge (F^{\ast }\Psi )(u_1,\dots ,u_{r+s})\end{array}$$

3.Deduce the transformation formula of coefficients of a $(r,s)$-tensor under the change
of local coordinates.

Proof:

Suppose $(U,\phi ;x^i)$ is a chart of $M$, $(r,s)$-tensor $\Phi$ can be locally represented:
$$\Phi (x)={\Phi }_{j_1\dots j_s}^{i_1\dots i_r}(x)\frac{\partial }{\partial x^{i_1}}\otimes \cdots \otimes \frac{\partial }{\partial x^{i_r}}\otimes d{x}^{j_1}\otimes \cdots \otimes d{x}^j$$

Thus for another chart $(V,\varphi ;x^{\prime i})$, we have the transformation formula on the intersection:

${\Phi }_{j_1\dots j_s}^{i_1\dots i_r}(x^{\prime })={\Phi }_{k_1\dots k_s}^{h_1\dots h_r}(x)\frac{\partial x^{\prime i_1}}{\partial x^{h_1}}\dots \frac{\partial x^{\prime i_r}}{\partial x^{h_r}}\frac{\partial x^{\prime k_1}}{\partial x^{\prime j_1}}\dots \frac{\partial x^{\prime k_s}}{\partial x^{\prime j_s}}$

4.Show that any $C^{\infty }(M)$-linear map
$$\tau :\underset{s}{\underbrace{{\Lambda }^1(M)\times \cdots \times {\Lambda }^1(M)}}\times \underset{r}{\underbrace{\mathcal{X}(M)\times \dots \mathcal{X}(M)}}\to C^{\infty }(M)$$
can be identified with a $(s,r)$-tensor field.

Proof:
We only show the result of $(0,1)$-tensor field, for the situation of $(r,s)$-tensor can be proved by Cartesian product of every vector field, and it’s too long to write completely.

Suppose we have a linear map $\tau :\mathcal{X}(M)\to C^{\infty }(M)$, s.t. $\forall v\in \mathcal{X}(M)$ and $\mu \in C^{\infty }(M)$ we have:
$$\tau (\mu \cdot v)=\mu \cdot \tau (v)$$

We need to find out a $(0,1)$-tensor field $\tilde{\tau }$ on $M$, s.t. $\forall x_0\in M$:
$$\tilde{\tau }(x_0)(X(x_0))=(\tau (X))(x_0),\forall X\in \mathcal{X}(M)$$

First, we hvae $\tau :\mathcal{X}(M)\to C^{\infty }(M)$ is local defined, $X,Y\in \mathcal{X}(M)$, and for open subset $U\subset M$, $X{|}_U=Y{|}_U$, we have $\tau (X){|}_U=\tau (Y){|}_U\in C^{\infty }(U)$:

By the partition of unity, $f\cdot (X-Y)$ is a zero vector field on $M$.

$\tau (0)=\tau (0)+\tau (0)\Rightarrow \tau (0)=0$ on zero vector field.

thus, $0=\tau (f\cdot (X-Y))=f\cdot (\tau (X)-\tau (Y))$, thus we do have $\tau (X){|}_U=\tau (Y){|}_U$.

Suppose $x_0\in (U;x^i)$ is a chart, and $v\in \mathcal{X}(U)$, thus
$$v=\sum _i{v}^i\frac{\partial }{\partial x^i}$$

thus, $\tau (v)={\sum }_i{v}^i\cdot \tau (\frac{\partial }{\partial x^i})$

Then, we can define $\tilde{\tau }(x_0)\in T_{x_0}^{\ast }M$ for $x_0\in M$:

$\forall v_0\in T_{x_0}M$, $v\in \mathcal{X}(U)$ is a extension of $v_0$ at $U$ whcih contains $x_0$ , s.t. $v(x_0)=v_0$, and
$$(\tilde{\tau }(x_0))(v_0)=(\tau (v))(x_0)$$

and $\tilde{\tau }(x_0):T_{x_0}M\to \mathbb{R}$ is linear, thus $\tilde{\tau }(x_0)\in T_{x_0}^{\ast }M$.

In $(U;x^i)$, we have:
$$\tilde{\tau }{|}_U=\sum _{i=1}^m\tilde{\tau }(\frac{\partial }{\partial x^i})d{x}^i$$

$$\tilde{\tau }(\frac{\partial }{\partial x^i})(x_0)=\tilde{\tau }(x_0)(\frac{\partial }{\partial x^i})=\tau (\frac{\partial }{\partial x^i})(x_0),\forall x_0\in U$$

Thus,
$$\tilde{\tau }(\frac{\partial }{\partial x^i})=\tau (\frac{\partial }{\partial x^i})\in C^{\infty }(U)$$

thus $\tilde{\tau }$ is cotangent vector field, and $\forall v\in \mathcal{X}(M)$, we have $\tilde{\tau }(v)=\tau (v)$