Multipass quantum process tomography

Area of Science:

• Quantum Information Science
• Quantum Computing
• Quantum Error Mitigation

Background:

• Quantum Process Tomography (QPT) is crucial for characterizing quantum operations.
• Standard QPT methods are susceptible to errors like state preparation and measurement (SPAM) errors, readout errors, and shot noise.
• Existing methods like Randomized Benchmarking provide only fidelity, not a complete process description.

Purpose of the Study:

• To introduce a novel QPT method enhancing precision and accuracy.
• To mitigate errors from SPAM, readout, and shot noise in quantum process characterization.
• To provide a comprehensive Pauli Transfer Matrix (PTM) rather than a single fidelity value.

Main Methods:

• Performing QPT on a sequence of multiple applications of the same quantum gate.
• Measuring the PTM of the multipass process via standard QPT.
• Deducing the single-process PTM using iterative and linearized approaches (Sylvester equation).
• Simulating and experimentally validating the method on IBM Quantum platforms.

Main Results:

• The proposed multipass QPT method significantly reduces SPAM, readout, and shot noise errors.
• Simulations confirm the efficiency of iterative and linearized deduction approaches.
• Experimental determination of the CNOT gate's PTM and fidelity on IBMQ_MANILA.
• The method yields the entire PTM, offering more information than fidelity-only metrics.

Conclusions:

• The multipass QPT approach offers superior accuracy and precision compared to standard QPT.
• This method effectively combats common noise sources in quantum computations.
• The technique provides a more complete characterization of quantum processes, essential for building reliable quantum computers.