When computing the gradients of a quantum neural network using the
parameter-shift rule, the cost function needs to be calculated twice for the
gradient with respect to a single adjustable parameter of the network. When the
total number of parameters is high, the quantum circuit for the computation has
to be adjusted and run for many times. Here we propose an approach to compute
all the gradients using a single circuit only, with a much reduced circuit
depth and less classical registers. We also demonstrate experimentally, on both
real quantum hardware and simulator, that our approach has the advantages that
the circuit takes a significantly shorter time to compile than the conventional
approach, resulting in a speedup on the total runtime.

本文提出了一种仅需一个量子电路就能计算量子神经网络梯度的方法，该方法减少了电路深度和经典寄存器数量，同时解决了解决原方法中参数数量较多时所面临的电路调整和重复运行的问题，并在实验中展示了快速编译的优势和缩短总运行时间的速度提升。

利用单个电路计算量子神经网络对所有参数的梯度

Computing the gradients with respect to all parameters of a quantum  neural network using a single circuit

The study of Variational Quantum Eigensolvers (VQEs) has been in the
spotlight in recent times as they may lead to real-world applications of
near-term quantum devices. However, their performance depends on the structure
of the used variational ansatz, which requires balancing the depth and
expressivity of the corresponding circuit. In recent years, various methods for
VQE structure optimization have been introduced but the capacities of machine
learning to aid with this problem has not yet been fully investigated. In this
work, we propose a reinforcement learning algorithm that autonomously explores
the space of possible ans{\"a}tze, identifying economic circuits which still
yield accurate ground energy estimates. The algorithm is intrinsically
motivated, and it incrementally improves the accuracy of the result while
minimizing the circuit depth. We showcase the performance of our algorithm on
the problem of estimating the ground-state energy of lithium hydride (LiH). In
this well-known benchmark problem, we achieve chemical accuracy, as well as
state-of-the-art results in terms of circuit depth.

本论文提出了一种基于强化学习的算法，能够自主探索合适的变分预测模型，同时最小化电路的深度并增加结果的精度，在锂氢分子比较经典的量子计算基准问题上取得了化学精度和最先进的电路深度优化结果。

强化学习用于变分量子电路架构优化

Reinforcement learning for optimization of variational quantum circuit  architectures

Quantum Amplitude Estimation (QAE) can achieve a quadratic speed-up for
applications classically solved by Monte Carlo simulation. A key requirement to
realize this advantage is efficient state preparation. If state preparation is
too expensive, it can diminish the quantum advantage. Preparing arbitrary
quantum states has exponential complexity with respect to the number of qubits,
thus, is not applicable. Currently known efficient techniques require problems
based on log-concave probability distributions, involve learning an unknown
distribution from empirical data, or fully rely on quantum arithmetic. In this
paper, we introduce an approach to simplify state preparation, together with a
circuit optimization technique, both of which can help reduce the circuit
complexity for QAE state preparation significantly. We demonstrate the
introduced techniques for a numerical integration example on real quantum
hardware, as well as for option pricing under the Heston model, i.e., based on
a stochastic volatility process, using simulation.

本研究提出了一种简化状态准备的方法以及电路优化技术，可以显著减少 QAE 状态准备的电路复杂度，并演示了在实际量子硬件上进行数字积分示例以及基于随机波动率过程的 Heston 模型的期权定价问题的技术应用。