Characterizing large-scale quantum computers via cycle benchmarking (1902.08543v1)

Abstract: Quantum computers promise to solve certain problems more efficiently than their digital counterparts. A major challenge towards practically useful quantum computing is characterizing and reducing the various errors that accumulate during an algorithm running on large-scale processors. Current characterization techniques are unable to adequately account for the exponentially large set of potential errors, including cross-talk and other correlated noise sources. Here we develop cycle benchmarking, a rigorous and practically scalable protocol for characterizing local and global errors across multi-qubit quantum processors. We experimentally demonstrate its practicality by quantifying such errors in non-entangling and entangling operations on an ion-trap quantum computer with up to 10 qubits, with total process fidelities for multi-qubit entangling gates ranging from 99.6(1)% for 2 qubits to 86(2)% for 10 qubits. Furthermore, cycle benchmarking data validates that the error rate per single-qubit gate and per two-qubit coupling does not increase with increasing system size.

• The paper introduces cycle benchmarking (CB) to isolate and quantify both local and global noise in large-scale quantum systems.
• It demonstrates the protocol on ion-trap systems, achieving process fidelities up to 99.6% for 2-qubit and 86% for 10-qubit operations.
• The method robustly decouples SPAM errors through random Pauli cycles, paving the way for scalable, fault-tolerant quantum computing.

Characterizing Large-Scale Quantum Computers via Cycle Benchmarking

The paper, "Characterizing large-scale quantum computers via cycle benchmarking," introduces an advanced method termed Cycle Benchmarking (CB), aimed at accurately characterizing noise in multi-qubit quantum processors. The motivation behind this paper arises from the challenge quantum computing currently faces: significant error rates that accumulate during the computation process, exacerbated by correlated and cross-talk errors, which current characterization techniques fail to adequately capture.

Key Contributions

1. Introduction of Cycle Benchmarking (CB): The CB protocol is a novel approach for effectively characterizing error rates across a large quantum system. It is designed to isolate and quantify both local and global noise mechanisms that impact quantum computing operations during these cycles. By focusing on a "cycle" of operations that act on the entire quantum register, CB provides a pragmatic solution to characterize errors in large-scale systems where the execution of complex computations amplifies these errors significantly.
2. Experimental Demonstration on Ion-Trap Quantum Systems: The authors experimentally validate the CB protocol using an ion-trap quantum computer, showcasing its scalability and effectiveness up to 10 qubits. They explore both local operations and global entangling gate processes, providing evidence of its practicality in real-world settings.
3. Insights into Process Fidelity: The paper reports high process fidelity rates, achieving up to 99.6% fidelity for 2-qubit systems with entangling multi-qubit operations, and 86% for 10 qubits. Furthermore, the paper finds that the error rate per single and two-qubit operations remains constant as the system size increases. This contradicts the notion that error rates would deteriorate with larger quantum computational systems due to added complexity and interaction.
4. Decoupling and Robustness to SPAM: Central to the protocol’s innovation is its robustness to State Preparation and Measurement (SPAM) errors. CB achieves this by applying random Pauli cycles, ensuring that the measurement and preparation errors are isolated from operational errors.

Theoretical and Practical Implications

The development of CB is founded upon a structured mathematical framework that verifies its scalability and accuracy. The paper includes rigorous theoretical backing that guarantees the extraction of process fidelity, even as system sizes increase. This is further underlined by the detailed derivations, such as the proofs regarding the independence of fidelity estimates from SPAM, ensuring that the results reported are not biased due to errors in state preparation and measurement.

Practically, the insights from CB have profound implications for the development of fault-tolerant quantum computing. The ability to effectively measure and understand errors in a scalable manner is critical for developing quantum error correction codes that can tolerate known errors, rather than underestimating the existing error rates due to uncharacterized noise.

Future Directions and Speculations

In the context of advancing quantum technologies, the implications of cycle benchmarking are vast. Future directions will likely explore the integration of CB into larger-than-10-qubit systems and other quantum architectures besides ion traps. The scalability of CB makes it a promising candidate for application in superconducting qubits and other platforms.

Furthermore, the extension of CB to non-Clifford operations is notable for broader applications, especially in algorithms requiring such operations as core components. Understanding error rates in these scenarios will be pivotal in comprehensive quantum algorithmic development. Additionally, future research could explore the adaptation of CB for dynamic noise environments and real-time error correction applications in quantum computing.

In conclusion, the paper lays a comprehensive foundation and provides substantial empirical evidence for the use of cycle benchmarking, positioning it as a substantial tool in the ongoing effort to make quantum computing more efficient and reliable. As the field evolves, methods like CB that offer granular fidelity insights will be indispensable in overcoming existing computational challenges.