arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.09028quant-phcs.LGstat.ML
收藏

具有优化随机特征的数据分类中的指数误差收敛:量子机器学习加速

Exponential Error Convergence in Data Classification with Optimized Random Features: Acceleration by Quantum Machine Learning

IQOQI Vienna, Austrian Academy of Sciences(奥地利科学院维也纳量子光学与量子信息研究所);Atominstitut, Technische Universität Wien(维也纳技术大学原子研究所);RIKEN AIP(理化学研究所人工智能研究中心)

Hayata Yamasaki, Sho Sonoda

摘要:

分类是机器学习中的一项常见任务。随机特征(RF)是基于核方法的可扩展学习算法的核心技术,最近提出的优化随机特征,根据模型和数据分布进行采样,可以显着减少并可证明最小化所需的特征数量。然而,现有的研究分类使用优化的RF遭受计算困难,在采样每个优化的RF,而且,它未能实现指数快速的误差收敛速度,其他国家的最先进的核方法可以实现在低噪声条件下。为了克服这些慢化,我们在这里构建了一个分类算法,通过量子机器学习(QML)加速优化的RF,并研究其运行时,以阐明整体优势。我们证明,我们的算法可以实现指数误差收敛下的低噪声条件下,即使与优化的RF,在同一时间,我们的算法可以利用的优势,显着减少的数量的功能,而不会由于QML的计算难度。这些结果发现了一个有前途的应用QML加速领先的基于核的分类算法,而不破坏其广泛的适用性和指数误差收敛速度。

英文摘要:

Classification is a common task in machine learning. Random features (RFs) stand as a central technique for scalable learning algorithms based on kernel methods, and more recently proposed optimized random features, sampled depending on the model and the data distribution, can significantly reduce and provably minimize the required number of features. However, existing research on classification using optimized RFs has suffered from computational hardness in sampling each optimized RF; moreover, it has failed to achieve the exponentially fast error-convergence speed that other state-of-the-art kernel methods can achieve under a low-noise condition. To overcome these slowdowns, we here construct a classification algorithm with optimized RFs accelerated by means of quantum machine learning (QML) and study its runtime to clarify overall advantage. We prove that our algorithm can achieve the exponential error convergence under the low-noise condition even with optimized RFs; at the same time, our algorithm can exploit the advantage of the significant reduction of the number of features without the computational hardness owing to QML. These results discover a promising application of QML to acceleration of the leading kernel-based classification algorithm without ruining its wide applicability and the exponential error-convergence speed.

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。

↑