arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.08606quant-phcs.LG
收藏

可训练量子机器学习中的任意多项式分离

Arbitrary Polynomial Separations in Trainable Quantum Machine Learning

Institute for Quantum Information and Matter(量子信息与物质研究所);Walter Burke Institute for Theoretical Physics, Caltech, Pasadena, CA, USA(沃尔特·伯克理论物理研究所,卡内基梅隆大学,帕萨迪纳,CA,USA);MIT Center for Theoretical Physics, Cambridge, MA, USA(麻省理工学院理论物理中心,剑桥,MA,USA);Department of Physics, CU Boulder, Boulder, CO, USA(科罗拉多大学博尔德分校物理系)

Eric R. Anschuetz, Xun Gao

AI总结:

本文提出了一种高效可训练的量子神经网络结构,实现了在经典序列建模任务中相对于经典神经网络的任意多项式内存分离。

摘要:

最近量子机器学习的理论结果展示了量子神经网络(QNNs)的表达能力与其可训练性之间的普遍权衡;作为这些结果的推论,实践中的指数级表达能力优势被认为在经典机器学习模型上是不可行的,因为此类QNNs的训练时间与模型规模呈指数关系。我们通过构建一个高效可训练的QNNs层次结构,绕过了这些负面结果。该结构展示了在执行经典序列建模任务时,相对于经典神经网络(包括最先进的模型,如Transformer)的无条件证明的任意常数度的多项式内存分离。此构造也是计算高效的,因为所引入的QNNs类的每个单元格仅具有常数门复杂度。我们表明,语义模糊性(非正式地,一种量化的语义模糊性概念)是表达能力分离的根源,这表明其他具有此性质的学习任务可能是量子学习算法的自然应用领域。

英文摘要:

Recent theoretical results in quantum machine learning have demonstrated a general trade-off between the expressive power of quantum neural networks (QNNs) and their trainability; as a corollary of these results, practical exponential separations in expressive power over classical machine learning models are believed to be infeasible as such QNNs take a time to train that is exponential in the model size. We here circumvent these negative results by constructing a hierarchy of efficiently trainable QNNs that exhibit unconditionally provable, polynomial memory separations of arbitrary constant degree over classical neural networks -- including state-of-the-art models, such as Transformers -- in performing a classical sequence modeling task. This construction is also computationally efficient, as each unit cell of the introduced class of QNNs only has constant gate complexity. We show that contextuality -- informally, a quantitative notion of semantic ambiguity -- is the source of the expressivity separation, suggesting that other learning tasks with this property may be a natural setting for the use of quantum learning algorithms.

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。

↑