Efficient Data Encoding and Decoding for Quantum Computing
Efficient Data Encoding and Decoding for Quantum Computing
复制标题
DOI：
10.1109/qce53715.2022.00110
复制
发表时间：
2022-09
期刊：
影响因子：
--
通讯作者：
Naveed Mahmud;M. Jeng;Md. Alvir Islam Nobel;Manu Chaudhary;S. Islam;David Levy;E. El-Araby
中科院分区：
文献类型：
--
作者：
Naveed Mahmud;M. Jeng;Md. Alvir Islam Nobel;Manu Chaudhary;S. Islam;David Levy;E. El-Araby
噪声中尺度量子（NISQ）器件面临许多关键挑战，这些挑战限制了它们在实际应用中的有用性。一个主要的挑战是经典到量子（C2 Q）数据编码，这需要特定的电路来进行量子态初始化，特别是对于I/O密集型应用。所需的状态初始化电路通常很复杂，并且违反去相干约束。量子计算机的另一个关键挑战是量子态读出或量子到经典（Q2 C）数据解码。Q2 C的一般方法涉及量子电路的重复采样，这通常会在总执行时间中产生显著的开销。在本文中，我们提出了量子算法的C2 Q数据编码和Q2 C数据解码的时间有效的方法。沿着分析了用于C2 Q数据编码的去相干优化电路。对于Q2 C，提出了一种基于量子小波变换的输出状态采样的新方法。所提出的方法进行了实验评估的国家的最先进的量子计算设备从IBM量子使用现实的多光谱数据。实验结果与我们的理论预期是一致的，并证实了我们提出的方法相比，现有技术的效率。更具体地说，我们提出的C2 Q方法证明了理论上电路深度减少了2倍，与最先进的方法相比，这导致了实验执行时间的改善，而我们的Q2 C方法实现了电路采样时间最多减少了89%。
Noisy Intermediate-Scale Quantum (NISQ) devices face many critical challenges that limit their usefulness for practical applications. A primary challenge is classical-to-quantum (C2Q) data encoding, which requires specific circuits for quantum state initialization, particularly for I/O intensive applications. The required state initialization circuits are often complex, and violate decoherence constraints. Another critical challenge for quantum computers is quantum state readout or quantum-to-classical (Q2C) data decoding. The general approach for Q2C involves repeated sampling of the quantum circuit, which often incurs significant overhead in the overall execution time. In this paper, we propose time-efficient methods for C2Q data encoding and Q2C data decoding for quantum algorithms. Decoherence optimized circuits for C2Q data encoding are presented along with analysis of their circuit depths. For Q2C, a novel approach based on more efficient sampling of the output state using the Quantum Wavelet Transform is proposed. The proposed methods are experimentally evaluated on a state-of-the-art quantum computing device from IBM Quantum using realistic multi-spectral data. Experimental results are consistent with our theoretical expectations and confirm the efficiency of our proposed methods compared to existing techniques. More specifically, our proposed C2Q method demonstrates a theoretical 2x reduction in circuit depth which resulted in improving the experimental execution time compared to the state-of-the-art, while our Q2C method achieved a maximum of 89% reduction in circuit sampling time.