Using a time dependent operator as a "control" term in the optimization module.

[Bill Huggins]

Is there a way to use a time dependent operator as one of the terms in the control Hamiltonian when doing optimization?

I am looking at a use case where my experimental control signal comes from using an arbitrary waveform generator to create the envelope for a high-frequency pure tone. It would seem natural to me to have the simple (but time dependent) pure tone be one of my "control" terms and the envelope generated by the AWG be the control pulse that is optimized.

Is this possible in qutip?

- Bill

[Alex Pitchford]

Yes, there is a demand for this feature. Not least our own research group here.

We have actually had a pull request made on the github for a time dependent drift. However, it stalled due to the disappearance of the developer before it reached a satisfactory state. I will try at raise it up my todo list.

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Alex Pitchford]

Having just read this again, I am not sure whether a time dependent drift is what you were asking for here. However, I think it should do what you are asking for here.

Anyway, I have just uploaded a PR for this:

https://github.com/qutip/qutip/pull/506

No example notebook yet, but I have done an example script that you can find at:

https://github.com/ajgpitch/qtrl-tutorial

[Bill Huggins]

Hi Alex,

The time dependent drift may be helpful in accomplishing the same task but what I was really hoping for was a time dependent driving term. That way I could optimize the amplitude of drives at several different frequencies without having a large number of free variables. I think that I may be able to use the CRAB algorithm's features to accomplish this for my case though.

Thank you for your help!

William Huggins

[Alex Pitchford]

Yes, I was also going to suggest CRAB as an option.

All the best

Alex

--

You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.