【摘要】 强大的计算能力是高效完成机器学习任务的有力保障。随着全球数据的飞速增长,更高的计算能力是机器学习领域的一个长期而紧迫的需求。利用量子计算的并行计算能力,量子机器学习算法相比于经典算法具有显著的速度优势,已经成为量子计算领域的研究热点。量子回归算法作为量子机器学习算法中的重要一类,近年来受到了广泛关注。本文综述了量子回归算法近年来的重要进展,包括量子线性回归、量子岭回归算法,并提出一个基于梯度下降法的量子逻辑回归算法。这些量子算法在合理的假设条件下相比经典算法有指数加速效果,展现出了量子计算的独特优势。更多还原

【Abstract】 The high computing performance is a powerful guarantee for efficient implement of machine learning tasks. With the rapid growth of global data, higher computing performance is a long-term and urgent demand in the field of machine learning. By means of the parallel computing capabilities of quantum computing, quantum machine learning algorithms, which is a research focus in the field of quantum computing, can get a significant speedup over classical algorithms. Quantum regression algorithms, as an important class of quantum machine learning algorithms, have received extensive attention in recent years. In this paper, we review the important progress of quantum regression algorithms in recent years, including quantum linear regression algorithms and quantum ridge regression algorithms. Also, we propose a quantum logic regression algorithm based on gradient descent. These quantum algorithms have exponential speedup over the classical algorithms under reasonable assumptions, which shows the unique advantages of quantum computing.更多还原