GST with assumed ideal gates

[ShuxiangCao]

Is your feature request related to a problem? Please describe. I am wondering if pyGSTi can support GST while some of the gates assumed to be known.

For example, on superconducting devices, the error of the virtual Z gate is much smaller compared to other gates, which can be approximately considered to be ideal. On the other hand, virtual Z gates are not physical gates, forcing them to be ideal can push all the errors to the physical gates and sometimes is more useful. Such configuration can also reduce the amount of computational complexity of GST and find the gate set with less effort of gauge optimization (the gauge would have to be commuted to virtual Z).

I tried to specify the operation to be a static arbitrary operator during the model construction, but it doesn't seem to be working.

Describe the solution you'd like

1. The gates would be considered ideal and not parametrized if it is specified as static operator during the model construction.
2. Add a gauge optimization option so that the gauge is parameterized to commute to certain gates.

May I know if this functionality can already be achieved with the current code, or if there is more effort required to develop it?

[sserita]

This is still an active area of development, but you should be able to do most of what you are describing with an OpFactory whose operations take a single argument (your Z phase) and return either a StaticArbitraryOp if you want it to be ideal or some other parameterization otherwise.

The OpFactory tutorial should be a good starting point for this, although there are more advanced options available than what is covered in this tutorial (for example, the continuously parameterized gates section of the ProcessorSpec tutorial.

[ShuxiangCao]

Thank you very much! I'll give it a try!

[sserita]

Closing this issue due to lack of activity and possible solution provided. Please open another issue if needed.