Mitigating noise in digital and digital-analog quantum computation

TL;DR

This paper demonstrates that digital-analog quantum computing (DAQC) offers superior noise resilience and higher fidelity compared to digital quantum computing in NISQ devices, especially when combined with error mitigation techniques.

Contribution

The study provides a comprehensive comparison of digital and DAQC paradigms under noise, highlighting DAQC's robustness and effectiveness in error mitigation for quantum algorithms.

Findings

DAQC outperforms digital quantum computing in fidelity under noise.

Zero-noise extrapolation significantly improves DAQC performance.

Fidelities above 0.95 achieved for 8-qubit systems with error mitigation.

Abstract

Noisy Intermediate-Scale Quantum (NISQ) devices lack error correction, limiting scalability for quantum algorithms. In this context, digital-analog quantum computing (DAQC) offers a more resilient alternative quantum computing paradigm that outperforms digital quantum computation by combining the flexibility of single-qubit gates with the robustness of analog simulations. This work explores the impact of noise on both digital and DAQC paradigms and demonstrates DAQC's effectiveness in error mitigation. We compare the quantum Fourier transform and quantum phase estimation algorithms under a wide range of single and two-qubit noise sources in superconducting processors. DAQC consistently surpasses digital approaches in fidelity, particularly as processor size increases. Moreover, zero-noise extrapolation further enhances DAQC by mitigating decoherence and intrinsic errors, achieving fidelities above 0.95 for 8 qubits, and reducing computation errors to the order of $10^{-3}$. These results establish DAQC as a viable alternative for quantum computing in the NISQ era.