Section 5.2

Boundedness and Operator Norms

[7/100%]
Show
LaTeX
Problem 5.2.1

Prove that B(X,Y)\mathcal{B}(X, Y) is a normed linear space. That is, prove Theorem 5.3.

Theorem 5.3: Let XX and YY be normed linear spaces. The collection B(X,Y)\mathcal{B}(X, Y) of all bounded linear operators T:X→YT: X \to Y, endowed with the operator norm, forms a normed linear space.

?
Problem 5.2.2

Show that the identity operator from (C([0,1]),∥⋅∥∞)\left(\mathcal{C}([0,1]),\left\| \cdot \right\|_{\infty }\right) to (C([0,1]),∥⋅∥1)\left(\mathcal{C}([0,1]),\left\| \cdot \right\|_{1}\right) is a bounded linear operator, but that the identity map from (C([0,1]),∥⋅∥1)\left(\mathcal{C}([0,1]),\left\| \cdot \right\|_{1}\right) to (C([0,1]),∥⋅∥∞)\left(\mathcal{C}([0,1]),\left\| \cdot \right\|_{\infty }\right) is unbounded. This phenomenon cannot happen on a Banach space; see Exercise 6.3.2.

?
Problem 5.2.3

Fix a continuous function ϕ:[0,1]→[0,1]\phi :[0,1] \rightarrow [0,1] and define the composition operator Cϕ:C([0,1])→C([0,1])C_{\phi }: C([0,1]) \rightarrow C([0,1]) by Cϕf(x)=f(ϕ(x))C_{\phi } f(x)=f(\phi (x)). Prove that CϕC_{\phi } is a bounded linear operator with ∥Cϕ∥≤1\left\| C_{\phi }\right\| \leq 1.

?
Problem 5.2.4

For {an}n=1∞∈ℓ∞\left\{ a_{n}\right\}_{n=1}^{\infty } \in \ell^{\infty } define

T(x)=∑n=1∞xnan, for all x={xn}n=1∞∈ℓ1 T(x)=\sum _{n=1}^{\infty } x_{n} a_{n}, \quad \text{ for all } x=\left\{ x_{n}\right\} _{n=1}^{\infty } \in \ell ^{1}

Show that TT is a bounded linear operator from ℓ1\ell^{1} to C\mathbb {C}, and compute ∥T∥\left\| T\right\|.

?
Problem 5.2.5

Consider an n×nn \times n matrix A=(aij)A=\left(a_{i j}\right) that satisfies aij=aji‾a_{i j}=\overline{a_{j i}} for all pairs of indices ii and jj. Such a matrix is called Hermitian. View AA as a linear operator Cn→Cn\mathbb {C}^{n} \rightarrow \mathbb {C}^{n}. Assume that Cn\mathbb {C}^{n} is endowed with the usual norm. Prove that ∥A∥=max⁡k∥λk∥\left\| A\right\| =\max_{k}\left\| \lambda_{k}\right\|, where λk\lambda_{k} are the eigenvalues of AA.

?
Problem 5.2.6

Consider

A=(abcd),a,b,c,d∈C A=\left(\begin{array}{ll} a & b \\ c & d \end{array}\right), \quad a, b, c, d \in \mathbb {C}

as a linear operator from C2\mathbb {C}^{2} to itself. The aim of this exercise is to show that the operator norm of AA depends on the choice of norm on C2\mathbb {C}^{2}.

?
(a)

Endow C2\mathbb {C}^{2} with the supremum norm, ∥⋅∥∞\left\| \cdot \right\|_{\infty }. Show that in this case, the norm of AA is given by

max⁡(∣a∣+∣b∣,∣c∣+∣d∣) \max (|a|+|b|,|c|+|d|)
(b)

Endow C2\mathbb {C}^{2} with the 1-norm, ∥⋅∥1\left\| \cdot \right\|_{1}. Show that in this case, the norm of AA is given by

max⁡(∣a∣+∣c∣,∣b∣+∣d∣) \max (|a|+|c|,|b|+|d|)
(c)

Endow C2\mathbb {C}^{2} with the 2-norm, ∥⋅∥2\left\| \cdot \right\|_{2}. Consider the matrix

A∗=(aˉcˉbˉdˉ) A^{*}=\left(\begin{array}{cc} \bar{a} & \bar{c} \\ \bar{b} & \bar{d} \end{array}\right)

and form the matrix AA∗A A^{*}. Show that in this case, the norm of AA is given by

[12(tr⁡(AA∗)+(tr⁡AA∗)2−4det⁡(AA∗))]12 \left[\frac{1}{2}\left(\operatorname {tr}\left(A A^{*}\right)+\sqrt{\left(\operatorname {tr} A A^{*}\right)^{2}-4 \operatorname {det}\left(A A^{*}\right)}\right)\right]^{\frac{1}{2}}

Use this to write down a formula for the norm of AA as a function of its entries a,b,c,da, b, c, d. Here "tr" denotes trace, and "det" denotes determinant.

(d)

Finally, compute these 3 operator norms of

A=(1101) A= \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix}
Problem 5.2.7

Let C1([0,1])C^{1}([0,1]) denote the collection of all continuous functions with continuous derivative (including one-sided derivatives at the endpoints).

?
(a)

Show that C1([0,1])C^{1}([0,1]), with the supremum norm, is a subspace of C([0,1])C([0,1]) but that it is not closed.

(b)

Prove that the differential operator

ddx:C′([0,1])→C([0,1]) \frac{d}{d x}: C^{\prime }([0,1]) \rightarrow C([0,1])

is linear but not bounded.