Let $(S(t))_{t>0}$ be a continuous operator from $L^2(0,1)$ to its self and Let $K$ be the operator $$\eqalign{ & K:{L^2}(0,1) \to {L^2}(0,1) \cr & f: \to (Kf)(x) = \int\limits_0^1 {k(s,x)S(s)f(s)ds} \cr} $$ where $$k \in {L^2}(0,1) \times {L^2}(0,1)$$ It is well known that if $S=I$ then $K$ is compact operator. What can I say about the compactness of $K$ is this case? Thank you.

  • 2
    $\begingroup$ I see no reason for $K$ even to be well-defined in general, forget about boundedness or compactness. You need to assume something else to make this construction meaningful (I interpret $S(s)f(s)$ as $(S(s)f)(s)$) $\endgroup$ – fedja Jun 9 at 0:46
  • $\begingroup$ I think you also need to be explicit as to exactly in what sense $S(t)$ is "continuous". Continuous for each $t$? A continuous curve in the space of (continuous?) operators (with what topology?) $\endgroup$ – Nate Eldredge Jun 9 at 2:56
  • $\begingroup$ In fact, $S(t)$ is C0_semogroup of operators on $L^2$ so it is continuous wth respect to its topology. $\endgroup$ – Gustave Jun 9 at 3:28

Your question must be reformulated to take into account the comments. Let us just say here that Hilbert-Schmidt operators (operators whose kernels are in $L^2((0,1)^2)$) make an ideal of the bounded operators so that composing a HS operator with a bounded operator gives a HS operator.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.