You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This project aims to investigate what happens in idling of qubits by conducting Interleaved Randomized Benchmarking (InterleavedRB) of gates with delay instructions using qiskit-experiment (c.f. qiskit-experiments/tutorials).
For example, we may examine how much the spectator qubit (q2) affects CNOT gate (on q0 and q1) fidelity by 3-qubit InterleavedRB of CNOT(q0,q1)+Delay(q2). And investigating if the fidelity can be estimated by the results from the normal 2-qubit InterleavedRB of CNOT(q1, q2) and 1-qubit InterleavedRB of Delay(q2).
The above is just an example. If you think of more interesting experimental scenario, let's pursue it!
Unfortunately, the current InterleavedRB has no full capability to conduct complicated experiments like the above experiment. So you'll have to upgrade it so that you can
interleave Delay instructions
create RB circuits without changing qubit layout during Clifford operations
compare multiple interleaved RB curves with the reference RB curve
etc.
Deliverables
A jupyter notebook investigating the results of RB of gates with delays.
Advanced: Pull requests to enhance the RB module in qiskit-experiments.
Mentors details
Mentor
Name: Toshinari Itoko
GitHub ID: @itoko-san
What they do: Qiskit developer (Terra/Aer), Researcher (circuit optimization/noise modeling)
Number of mentees
1 or 2
Type of mentees
Required:
Interest in Randomized benchmarking and qiskit-experiments
Python basics (including jupyter notebook)
Nice to have:
Good knowledge of Qiskit transpiler would be helpful
The text was updated successfully, but these errors were encountered:
Interleaved benchmarking module in qiskit-experiments supported interleaving gates and quantumcircuit but not instruction such as delay. We fixed code to interleave delay as instruction.
made a development environment and install qiskit-experiment, qiskit-terra in editable mode
write a testcode to confirm the existing qiskit-experiments fails interleaving delay element and quantumcircuit with delay with unittest
fix the code
to make 'delay' regarded as instruction
With the new delay element, we want to check how well the system keep the qubit state especially when there is no operation. When delay IRB is run on more than 3 qubit, the effect of spectator qubit can also be recognized.
Delay can take an arbitrary duration than I Gate which depends on backend properties.
Finally, we can evaluate how well the theory explains the real world by comparing the infidelity of delay which is an experimental value excluding gate dependent error and coherence limited EPG(error per gate) which is a theoretical one .
Description
This project aims to investigate what happens in idling of qubits by conducting Interleaved Randomized Benchmarking (InterleavedRB) of gates with delay instructions using qiskit-experiment (c.f. qiskit-experiments/tutorials).
For example, we may examine how much the spectator qubit (q2) affects CNOT gate (on q0 and q1) fidelity by 3-qubit InterleavedRB of CNOT(q0,q1)+Delay(q2). And investigating if the fidelity can be estimated by the results from the normal 2-qubit InterleavedRB of CNOT(q1, q2) and 1-qubit InterleavedRB of Delay(q2).
The above is just an example. If you think of more interesting experimental scenario, let's pursue it!
Unfortunately, the current InterleavedRB has no full capability to conduct complicated experiments like the above experiment. So you'll have to upgrade it so that you can
etc.
Deliverables
A jupyter notebook investigating the results of RB of gates with delays.
Advanced: Pull requests to enhance the RB module in qiskit-experiments.
Mentors details
Number of mentees
1 or 2
Type of mentees
The text was updated successfully, but these errors were encountered: