Leakage Benchmarking for Universal Gate Sets.

Area of Science:

• Quantum Information Science
• Quantum Computing
• Quantum Error Correction

Background:

• Leakage errors, the loss of quantum information to extraneous states, are a significant challenge in quantum computing.
• Existing methods for benchmarking leakage are often sensitive to noise and limited to single qubits.
• Developing robust leakage detection and correction is crucial for advancing fault-tolerant quantum computation.

Purpose of the Study:

• To propose an efficient and accurate framework for measuring leakage rates in multi-qubit quantum systems.
• To develop a protocol less sensitive to state preparation and measurement (SPAM) noise compared to existing methods.
• To enable benchmarking of multi-qubit leakages, a capability not previously achieved.

Main Methods:

• Introduction of Leakage Randomized Benchmarking (LRB) for quantifying leakage in quantum systems.
• Development of an interleaved variant (iLRB) to benchmark average leakage rates of multi-qubit gates under noise assumptions.
• Experimental demonstration of iLRB on two-qubit gates using flux tuning, with analysis against leakage models.

Main Results:

• LRB demonstrates improved insensitivity to SPAM noise and fewer gate set assumptions.
• LRB successfully benchmarks leakage in multi-qubit systems for the first time.
• Numerical experiments show good agreement between iLRB protocol results and theoretical leakage models.

Conclusions:

• The proposed LRB and iLRB protocols offer efficient and accurate methods for leakage characterization in quantum systems.
• These protocols advance the ability to detect and mitigate leakage errors, crucial for scalable quantum computing.
• The feasibility of LRB and iLRB paves the way for more reliable quantum hardware development and error correction strategies.