Help to transpile a parameterized circuit with more than 3 controls

[pdhoolia]

I can't transpile a parameterized circuit with more than 3 controls

from qiskit.compiler import transpile
from qiskit.circuit import QuantumCircuit, ParameterVector

p = ParameterVector('p',4)

pqc = QuantumCircuit(4)
for i in range(4):
    pqc.ry(p[i],i)

pqc = pqc.control(4,0,annotated=True)
pqc = transpile(pqc, basis_gates=['u1', 'u2', 'u3', 'cx'], optimization_level=3)

With the code above I got the error: QiskitError: 'Cannot synthesize MCRY with unbound parameter: p[0].'

Is there a way I can work around this. Like call assign_parameters prior to transpilation?

[pdhoolia]

The error you're encountering is due to the fact that Qiskit currently cannot synthesize multi-controlled rotation gates (like MCRY) when the parameters are unbound. This is because the synthesis of such gates requires concrete numerical values to generate the appropriate gate decomposition.

To work around this issue, you should assign the parameters before transpiling the circuit. This means using the assign_parameters method on your quantum circuit to bind the parameters to specific values before you attempt to transpile it. Here's how you can modify your code to do this:

1. Assign parameters before calling transpile.
2. Ensure that your circuit is fully parameterized with numerical values.

Here's the updated code:

from qiskit.compiler import transpile
from qiskit.circuit import QuantumCircuit, ParameterVector

# Define your parameters
p = ParameterVector('p', 4)

# Create a parameterized quantum circuit
pqc = QuantumCircuit(4)
for i in range(4):
    pqc.ry(p[i], i)

# Assign specific values to the parameters
param_values = {p[i]: 0.5 for i in range(4)}  # Example: binding all parameters to 0.5
pqc = pqc.assign_parameters(param_values)

# Control the circuit with 4 controls
pqc = pqc.control(4, 0, annotated=True)

# Transpile the circuit
pqc = transpile(pqc, basis_gates=['u1', 'u2', 'u3', 'cx'], optimization_level=3)

print(pqc)

Explanation:

• Parameter Assignment: Before attempting to transpile, we bind the parameters to specific numerical values using assign_parameters. This ensures that the transpiler can synthesize the multi-controlled rotation gate.

• Control and Transpile: After assigning parameters, the circuit is controlled and then transpiled. This sequence ensures that the transpiler works with a concrete, non-symbolic circuit.

Additional Insights:

• Parameterized Gates: If you need to work with parameterized circuits at runtime, consider using a backend that supports runtime parameter binding. However, for transpilation purposes, parameters must be bound to numerical values.

• Qiskit Updates: Keep an eye on updates to Qiskit, as future versions may include enhanced support for parameterized synthesis of multi-controlled gates.

This approach should resolve the error you're encountering and allow you to successfully transpile your circuit.