Noise Resilient Compilation Policies for Quantum Approximate Optimization Algorithm
Noise Resilient Compilation Policies for Quantum Approximate Optimization Algorithm
复制标题
量子近似优化算法的抗噪编译策略
DOI：
10.1145/3400302.3415745
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Mahabubul Alam, Abdullah Ash-Saki
中科院分区：
文献类型：
--
作者：
Mahabubul Alam, Abdullah Ash-Saki
量子近似优化算法（QAOA）是一种很有前途的量子-经典混合算法，用于求解含噪声量子器件的组合优化问题。在用于QAOA的量子电路中使用的多量子位CPHASE门是可交换的，即，可以改变门的顺序而不改变输出状态。这种重新排序导致并行执行更多的门和更少数量的附加SWAP门来编译QAOA电路，从而导致更低的电路深度和门计数。门的数量越少，通常表明门错误的累积越少，而电路深度的减少意味着量子位的退相干时间越少。然而，近期的量子器件在门成功概率方面表现出显着的变化。变量感知编译策略（即将大多数门操作放在具有较高门成功概率的量子位上）可以提高硬件上成功执行程序的概率。QAOA电路的更大灵活性为QAOA定制的编译策略提供了更好的优化范围。本文提出了一个参数的编译策略，以利用QAOA电路的独特特性，以及噪声设备的变化意识。针对ibmq_16_melbourne上的一组QAOA-MaxCut问题，我们提出了两种方法-变分感知量子比特放置（VQP）和变分感知迭代映射（Vim），它们可以显著提高电路成功概率（平均为1.8408倍）。
Quantum approximate optimization algorithm (QAOA) is a promising quantum-classical hybrid algorithm to solve hard combinatorial optimization problems using noisy quantum devices. The multiqubit CPHASE gates used in the quantum circuit for QAOA are commutative i.e., the order of the gates can be altered without changing the output state. This re-ordering leads to the execution of more gates in parallel and a smaller number of additional SWAP gates to compile the QAOA circuit resulting in lower circuit-depth and gate-count. A less number of gates generally indicates a lower accumulation of gate-errors, and a reduced circuit-depth means less decoherence time for the qubits. However, near-term quantum devices exhibit significant variations in the gate success probabilities. Variation-aware compilation policies (i.e. putting most gate operations on qubits with higher gate success probabilities) can enhance the probability of successful program execution on the hardware. The greater flexibility of QAOA-circuits offer better scope of optimization with QAOA-tailored compilation policies. This paper presents an argument for compilation policies to exploit the unique characteristics of QAOA-circuits alongside the variation-awareness of the noisy devices. We present two procedures - variation-aware qubit placement (VQP) and variation-aware iterative mapping (VIM) that can improve the circuit success probability quite significantly (≈8.408X on average) for a set of QAOA-MaxCut problems on ibmq_16_melbourne.
登录
查看更多内容
MUQUT：NISQ 计算机上的多约束量子电路映射：特邀论文
DOI：
10.1109/iccad45719.2019.8942132
发表时间：
2019
期刊：
2019 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)
影响因子：
--
作者：
Debjyoti Bhattacharjee;Abdullah Ash;M. Alam;A. Chattopadhyay;Swaroop Ghosh
通讯作者：
Swaroop Ghosh
影响因子：
5.5
作者：
Grimsley, Harper R.;Claudino, Daniel;Mayhall, Nicholas J.
通讯作者：
Mayhall, Nicholas J.
影响因子：
6.7
作者：
D. Venturelli;M. Do;E. Rieffel;J. Frank
通讯作者：
D. Venturelli;M. Do;E. Rieffel;J. Frank
影响因子：
6.4
作者：
Preskill, John
通讯作者：
Preskill, John
影响因子：
12.5
作者：
Zhou, Leo;Wang, Sheng-Tao;Lukin, Mikhail D.
通讯作者：
Lukin, Mikhail D.