arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2109.11676quant-phcs.LGstat.ML
收藏

量子神经网络中的过参数化理论

Theory of overparametrization in quantum neural networks

Los Alamos National Laboratory(洛斯阿拉莫斯国家实验室);Universidad de Buenos Aires(布宜诺斯艾利斯大学);Barcelona Supercomputing Center(巴塞罗那超级计算中心);Instituto de Física Teórica, UAM-CSIC(UAM-CSIC理论物理研究所)

Martin Larocca, Nathan Ju, Diego García-Martín, Patrick J. Coles, M. Cerezo

AI总结:

本文严格分析了具有周期性结构的量子神经网络(QNN)中的过参数化现象,证明了李代数维数是临界参数数量及矩阵最大秩的上界,揭示了过参数化引发改善可训练性的计算相变,并将其与QNN容量相联系。

摘要:

用量子神经网络(QNNs)实现量子优势的前景令人振奋。理解QNN的特性(如参数数量$M$)如何影响损失景观,对于设计可扩展的QNN架构至关重要。在此,我们严格分析了具有周期性结构的QNNs中的过参数化现象。我们将过参数化定义为QNN拥有超过临界参数数量$M_c$的区间,该数量允许其探索状态空间中的所有相关方向。我们的主要结果表明,由QNN生成元获得的李代数的维数是$M_c$的上界,也是量子Fisher信息和Hessian矩阵能达到的最大秩的上界。欠参数化的QNNs在损失景观中存在伪局部极小值,当$M\geq M_c$时这些极小值开始消失。因此,过参数化的出现对应于一种计算相变,此时QNN的可训练性通过更有利的景观得到极大改善。随后,我们将过参数化的概念与QNN容量联系起来,使得当QNN过参数化时,其容量达到最大可能值。我们针对本征求解器、编译和自动编码应用运行了数值模拟,以展示过参数化的计算相变。我们注意到,我们的结果同样适用于变分量子算法和量子最优控制。

英文摘要:

The prospect of achieving quantum advantage with Quantum Neural Networks (QNNs) is exciting. Understanding how QNN properties (e.g., the number of parameters $M$) affect the loss landscape is crucial to the design of scalable QNN architectures. Here, we rigorously analyze the overparametrization phenomenon in QNNs with periodic structure. We define overparametrization as the regime where the QNN has more than a critical number of parameters $M_c$ that allows it to explore all relevant directions in state space. Our main results show that the dimension of the Lie algebra obtained from the generators of the QNN is an upper bound for $M_c$, and for the maximal rank that the quantum Fisher information and Hessian matrices can reach. Underparametrized QNNs have spurious local minima in the loss landscape that start disappearing when $M\geq M_c$. Thus, the overparametrization onset corresponds to a computational phase transition where the QNN trainability is greatly improved by a more favorable landscape. We then connect the notion of overparametrization to the QNN capacity, so that when a QNN is overparametrized, its capacity achieves its maximum possible value. We run numerical simulations for eigensolver, compilation, and autoencoding applications to showcase the overparametrization computational phase transition. We note that our results also apply to variational quantum algorithms and quantum optimal control.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑