Approximate quantum random access memory architectures
K. Phalak, J. Li, S. Ghosh · 2022 · arXiv
| 编号 QR020 | 主题 综述与物理基础 | 链接 arXiv:2210.14804 |
摘要
摘要保留原文（英文）
摘要取自 arXiv / 出版方元数据，未作翻译，以保证与原文一致。
Abstract—Quantum supremacy in many applications using well-known quantum algorithms rely on availability of data in quantum format. Quantum Random Access Memory (QRAM), an equivalent of classical Random Access Memory (RAM), fulfills this requirement. However, the existing QRAM proposals either require qutrit technology and/or incur access challenges. We propose an approximate Parametric Quantum Circuit (PQC) based QRAM which takes address lines as input and gives out the corresponding data in these address lines as the output. We present two applications of the proposed PQC-based QRAM namely, storage of binary data and storage of machine learning (ML) dataset for classification. For the machine learning task, we perform binary classification using three different setups, (a) QNN with QRAM (QRAM + QNN), (b) QNN with normal data embedding and (c) Fully Connected Neural N k (FCNN) W h h QRAM QNN ( i h l ifi i ) f i b th h d the other two setups which converge at around 10th and 15th epochs. The loss for QRAM + QNN (0.53) is less than FCNN setup (0.89) and comparable to QNN with normal embedding (0.48). For the storage of binary data, we evaluate Hamming Distance (HD) and percentage correct prediction metrics to quantify the performance. We observe an increase in HD from 0 to 3.97 and decrease in percentage of correct predictions from 100% to 0.58% as we increase the number of address and data lines from 2 to 9. We propose agglomerative clustering
本站词条在引用该文献时，会以 wikilink（sources/QR020）链接到本页；右侧「反向链接」面板自动列出所有引用它的词条。