- The paper reports the first experimental realization of repeated error correction in a distance-3 surface code on a 17-qubit superconducting quantum processor.
- The experiment utilized a 17-qubit configuration to implement repeated QEC cycles, achieving an approximate 20% reduction in logical error rates using a minimum weight perfect matching algorithm.
- This work demonstrates a crucial step towards scalable fault-tolerant quantum computing but highlights challenges in fidelity limited by readout error and coherence times.
Realization of an Error-Correcting Surface Code with Superconducting Qubits
In the pursuit of scalable fault-tolerant quantum computation, the implementation of quantum error correction (QEC) techniques represents a pivotal advancement. Among various QEC codes, the surface code stands out due to its high error threshold and implementation compatibility with two-dimensional qubit arrays—such as those used in superconducting and quantum dot systems. This paper reports the experimental realization of a distance-3 error-correcting surface code using a 17-qubit configuration on the Zuchongzhi 2.1 superconducting quantum processor. This experiment marks the first occasion of achieving repeated error correction in a surface code, thereby progressing towards the ultimate goal of fault-tolerant quantum computing.
A significant advancement over prior experiments, which were limited to error detection, is achieved by implementing repeated error correction cycles. Through encoding a logical qubit in a distance-3 code, the authors demonstrate an approximate 20% reduction in logical error rates post-correction using an error-correcting algorithm based on minimum weight perfect matching. The surface code architecture utilizes 9 data qubits and 8 ancilla qubits, operating on a two-dimensional grid with nearest-neighbor interactions. This design allows for the execution of repeated QEC cycles, which involves stabilizer measurements performed via ancilla qubits to detect and analyze errors.
The authors performed an in-depth calibration and evaluation of the quantum system, achieving single and two-qubit gate errors of 0.098% and 1.035%, respectively. These results were accrued over real-time circuit sampling tasks with a notable cross-entropy benchmarking fidelity of 0.021. Despite the promising results, fidelity remains a challenge, primarily constrained by the readout error rate and coherence times, issues exacerbated by the current measurement duration. Potential solutions include optimization in feedback control and further reduction of gate error rates through advanced fabrication techniques.
The implications of this work are profound. A successfully implemented repeated QEC operation suggests that building larger-scale surface codes, which require even greater qubit counts and further error correction layers, could one day enable practical quantum algorithms to be executed reliably on noisy quantum devices. Future directions include scaling the surface code to higher dimensions and enhancing the quantum processors' coherence and operation fidelities further, thereby reducing logical error rates to levels surpassing physical error rates. Such advancements will require concurrent improvements in circuit design, system integration, and error correction algorithms.
In summary, this paper successfully transitions from previous detection-only capabilities to repeated QEC cycles, highlighting both the practical and theoretical strides necessary for quantum computing development. As quantum processors become more sophisticated, this foundational work on error correction through surface codes will serve as a vital component in advancing towards robust and scalable quantum computing architectures.