Section II.4

Compact Operators

[14/0%]
LaTeX
Problem II.4.1

Prove Proposition 4.2(c): If A∈B(H)A \in \mathcal{B}(\mathcal{H}), B∈B(K)B \in \mathcal{B}(\mathcal{K}), and T∈B0(H,K)T \in \mathcal{B}_{0}(\mathcal{H}, \mathcal{K}), then TATA and BT∈B0(H,K)BT \in \mathcal{B}_{0}(\mathcal{H}, \mathcal{K}).

?
Problem II.4.2

Show that every operator of finite rank is compact.

?
Problem II.4.3

If T∈B00(H,K)T \in \mathcal{B}_{00}(\mathcal{H}, \mathcal{K}), show that T∗∈B00(K,H)T^{*} \in \mathcal{B}_{00}(\mathcal{K}, \mathcal{H}) and dim⁡(ran⁡T)=dim⁡(ran⁡T∗)\dim (\operatorname {ran} T) = \dim (\operatorname {ran} T^{*}).

?
Problem II.4.4

Show that an idempotent is compact if and only if it has finite rank.

?
Problem II.4.5

Show that no nonzero multiplication operator on L2(0,1)L^{2}(0,1) is compact.

?
Problem II.4.6

Show that if T:H→KT: \mathcal{H} \rightarrow \mathcal{K} is a compact operator and {en}\left\{ e_{n}\right\} is any orthonormal sequence in H\mathcal{H}, then ∥Ten∥→0\left\| Te_{n}\right\| \rightarrow 0. Is the converse true?

?
Problem II.4.7

If TT is compact and M\mathcal{M} is an invariant subspace for TT, show that T∣MT|\mathcal{M} is compact.

?
Problem II.4.8

If h,g∈Kh, g \in \mathcal{K}, define T:H→KT: \mathcal{H} \rightarrow \mathcal{K} by Tf=⟨f,h⟩gTf = \left\langle f, h \right\rangle g. Show that TT has rank 11 [that is, dim⁡(ran⁡T)=1\dim (\operatorname {ran} T) = 1]. Moreover, every rank 11 operator can be so represented. Show that if TT is a finite rank operator, then there are orthonormal vectors e1,…,ene_{1}, \ldots , e_{n} and vectors g1,…,gng_{1}, \ldots , g_{n} such that Th=∑j=1n⟨h,ej⟩gjTh = \sum_{j=1}^{n} \left\langle h, e_{j} \right\rangle g_{j} for all hh in H\mathcal{H}. In this case show that TT is normal if gj=λjejg_{j} = \lambda_{j} e_{j} for some scalars λ1,…,λn\lambda_{1}, \ldots , \lambda_{n}. Find σp(T)\sigma_{p}(T).

?
Problem II.4.9

Show that a diagonalizable operator is normal.

?
Problem II.4.10

Verify the statements in Example 4.10: Let AA be the diagonalizable operator in Proposition 4.6 (that is, H\mathcal{H} is a separable Hilbert space with basis {en}\left\{ e_{n}\right\}; {αn}⊆F\left\{ \alpha_{n}\right\} \subseteq \mathbb {F} with M=sup⁡{∣αn∣:n≥1}<∞M = \sup \left\{ \left|\alpha_{n}\right|: n \geq 1\right\} < \infty; Aen=αnenAe_{n} = \alpha_{n} e_{n} for all nn). Then σp(A)={α1,α2,…}\sigma_{p}(A) = \left\{ \alpha_{1}, \alpha_{2}, \ldots \right\}. If α∈σp(A)\alpha \in \sigma_{p}(A), let Jα={j∈N:αj=α}J_{\alpha } = \left\{ j \in \mathbb {N}: \alpha_{j} = \alpha \right\}. Then hh is an eigenvector for α\alpha if and only if h∈∨{ej:j∈Jα}h \in \vee \left\{ e_{j}: j \in J_{\alpha }\right\}.

?
Problem II.4.11

Verify the statement in Example 4.11: The Volterra operator has no eigenvalues.

?
Problem II.4.12

Verify the statement in Example 4.12: Let h∈H=LC2(−π,π)h \in \mathcal{H} = L_{\mathbb {C}}^{2}(-\pi , \pi ) and define K:H→HK: \mathcal{H} \rightarrow \mathcal{H} by (Kf)(x)=(2π)−1/2∫−ππh(x−y)f(y) dy(Kf)(x) = (2\pi )^{-1/2} \int_{-\pi }^{\pi } h(x-y) f(y) \, dy. If λn=(2π)−1/2∫−ππh(x)exp⁡(−inx) dx=h^(n)\lambda_{n} = (2\pi )^{-1/2} \int_{-\pi }^{\pi } h(x) \exp (-inx) \, dx = \hat{h}(n), the nnth Fourier coefficient of hh, then Ken=λnenKe_{n} = \lambda_{n} e_{n}, where en(x)=(2π)−1/2exp⁡(−inx)e_{n}(x) = (2\pi )^{-1/2} \exp (-inx). (Note that the operator KK in this example is diagonalizable.)

?
Problem II.4.13

If Tn∈B(Hn)T_{n} \in \mathcal{B}(\mathcal{H}_{n}), n≥1n \geq 1, with sup⁡n∥Tn∥<∞\sup_{n} \left\| T_{n}\right\| < \infty and T=⊕n=1∞TnT = \oplus_{n=1}^{\infty } T_{n} on H=⊕n=1∞Hn\mathcal{H} = \oplus_{n=1}^{\infty } \mathcal{H}_{n}, show that TT is compact if and only if each TnT_{n} is compact and ∥Tn∥→0\left\| T_{n}\right\| \rightarrow 0.

?
Problem II.4.14

In Lemma 4.8, show that if L2(X,Ω,μ)L^{2}(X, \Omega , \mu ) is separable, then {ϕij}\left\{ \phi_{ij}\right\} is a basis for L2(X×X,Ω×Ω,μ×μ)L^{2}(X \times X, \Omega \times \Omega , \mu \times \mu ). What if L2(X,Ω,μ)L^{2}(X, \Omega , \mu ) is not separable?

Lemma 4.8: If {ei:i∈I}\left\{ e_{i}: i \in I\right\} is a basis for L2(X,Ω,μ)L^{2}(X, \Omega , \mu ) and

ϕij(x,y)=ej(x)ei(y)‾ \phi _{ij}(x,y) = e_{j}(x) \overline{e_{i}(y)}

for i,ji,j in II and x,yx,y in XX, then {ϕij:i,j∈I}\left\{ \phi_{ij}: i,j \in I\right\} is an orthonormal set in L2(X×X,Ω×Ω,μ×μ)L^{2}(X \times X, \Omega \times \Omega , \mu \times \mu ). If kk and KK are as in Proposition 4.7, then ⟨k,ϕij⟩=⟨Kej,ei⟩\left\langle k, \phi_{ij} \right\rangle = \left\langle Ke_{j}, e_{i} \right\rangle.

?