Ker(I-L) Finite-Dimensional: Proof

In summary: By the Riesz lemma, a compact operator on a compact space has finite-dimensional range. So, $\ker(I-L)$ must also be finite-dimensional. In summary, we have shown that $\ker(I-L)$ is finite-dimensional.
  • #1
Fermat1
187
0
Let L be an compact operator on a compact space K , and Let I be the identity on K.
Show that Ker(I-L) is finite-dimensional.

My attempt: Let $x_{n}$ be a sequence in the unit ball. K is compact so $(I(x_{n}))=(x_{n})$ has a convergent subsequence and L is compact operator so $L(x_{n})$ has a convergent subsequence. Thus, $(I-L)(x_{n})$ has a convergent subsequence so it is compact. $I-L$ restricted to the kernel is the identity, so since the identity is compact iff the space is finite dimensional, the kernel is finite dimensional. Is this correct?
 
Physics news on Phys.org
  • #2


Yes, your approach is correct. Here is a more detailed explanation:

Since $K$ is compact, it is also bounded. This means that for any sequence $(x_n)$ in $K$, there exists a convergent subsequence $(x_{n_k})$ such that $\lim_{k\to\infty} x_{n_k}=x\in K$.

Now, since $L$ is a compact operator, it maps any bounded sequence in $K$ to a sequence with a convergent subsequence. So, $(L(x_{n_k}))$ has a convergent subsequence, say $(L(x_{n_{k_j}}))$. Let $\lim_{j\to\infty} L(x_{n_{k_j}})=y\in K$.

Since $I$ is the identity operator, we have $I(x_{n_{k_j}})=x_{n_{k_j}}$ for all $j$. So, $(I(x_{n_{k_j}}))$ is a subsequence of $(x_{n_k})$ and hence, it also converges to $x$.

Now, let $x_{n_{k_j}}\in\ker(I-L)$. Then, $(I-L)(x_{n_{k_j}})=I(x_{n_{k_j}})-L(x_{n_{k_j}})=x_{n_{k_j}}-L(x_{n_{k_j}})$. Since $(x_{n_{k_j}})$ and $(L(x_{n_{k_j}}))$ both converge to $x$ and $y$, respectively, we have $\lim_{j\to\infty} (x_{n_{k_j}}-L(x_{n_{k_j}}))=x-y$.

Since $K$ is compact, it is also closed. So, $x-y\in K$. But $x-y$ is a limit point of the sequence $(x_{n_{k_j}}-L(x_{n_{k_j}}))$ which is contained in $\ker(I-L)$. So, $x-y\in\ker(I-L)$. This shows that every limit point of $\ker(I-L)$ is also contained in $\ker(I-L)$. Hence, $\ker(I-L)$ is closed.

Now, since $\ker(I-L)$ is a closed subset of the compact set $K$, it
 

FAQ: Ker(I-L) Finite-Dimensional: Proof

What does "Ker(I-L) Finite-Dimensional: Proof" mean?

"Ker(I-L) Finite-Dimensional: Proof" refers to a mathematical proof that demonstrates the finite-dimensionality of the kernel of a linear transformation, where I is the identity transformation and L is the linear transformation in question.

Why is this proof important?

This proof is important because it helps us understand the behavior of linear transformations and their kernels, which are fundamental concepts in linear algebra. It also has many applications in fields such as physics, engineering, and computer science.

What is the significance of the identity transformation in this proof?

The identity transformation, denoted by I, is important because it represents the simplest type of linear transformation - one that preserves the original vector. In this proof, it allows us to focus on the behavior of the linear transformation L without any additional complications.

Can you give an example of how this proof is used in real life?

One example of how this proof is used in real life is in computer graphics, where linear transformations are used to manipulate 3D objects. By understanding the finite-dimensionality of the kernel, we can ensure that the transformed object remains bounded and does not expand infinitely.

Is this proof limited to finite-dimensional spaces?

Yes, this proof is limited to finite-dimensional spaces. The concept of a kernel is not well-defined in infinite-dimensional spaces, so this proof cannot be applied in those cases.

Similar threads

Back
Top