Demonstrating quantum error mitigation on logical qubits

Area of Science:

• Quantum Computing
• Quantum Error Correction
• Quantum Information Science

Background:

• Quantum computing faces significant challenges from qubit noise, hindering early applications.
• Mitigating post-correction logical failures is crucial for advancing fault-tolerant quantum computing.

Purpose of the Study:

• To propose and experimentally validate zero-noise extrapolation for error correction circuits.
• To demonstrate a practical method for suppressing logical errors in superconducting quantum processors.

Main Methods:

• Applied zero-noise extrapolation, a quantum error mitigation technique.
• Amplified noise on physical qubits to establish a predictable polynomial dependence on noise strength.
• Utilized polynomial extrapolation to mitigate logical errors based on code distance.

Main Results:

• Achieved a universal reduction in logical errors across diverse quantum circuits.
• Demonstrated effectiveness in fault-tolerant circuits, including repetition and surface codes.
• Observed sustained performance with increased circuit depth in multi-round error correction.

Conclusions:

• Zero-noise extrapolation is a viable technique for mitigating logical errors in quantum error correction.
• This method offers a practical pathway to reliable quantum computing in the early fault-tolerant era.
• The approach shows promise for scaling quantum computations with improved fidelity.