arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.09328quant-phcs.AIcs.LGstat.ML
收藏

量子策略梯度算法

Quantum policy gradient algorithms

University of Innsbruck(因斯布鲁克大学);QuSoft(量子软件研究中心);University of Amsterdam(阿姆斯特丹大学);Leiden University(莱顿大学)

Sofiene Jerbi, Arjan Cornelissen, Māris Ozols, Vedran Dunjko

AI总结:

针对大状态与动作空间下强化学习量子算法适用性有限的问题,设计了利用环境量子交互训练前沿策略的量子算法,仅当策略满足正则性条件时可实现样本复杂度的二次加速,且参数化量子电路衍生的策略符合该条件,凸显全量子强化学习框架的价值。

摘要:

理解机器学习任务中量子数据访问的能力与局限性,是评估量子计算在人工智能领域潜力的首要前提。已有研究表明,在对强化学习环境具备量子访问权限的情况下,学习过程可实现加速。但目前该场景下的量子算法适用性仍十分有限,尤其是在状态和动作空间规模较大的环境中。本研究设计了多种量子算法,通过利用与环境的量子交互来训练前沿的强化学习策略。不过,仅当被训练的策略满足某些正则性条件时,这些算法在样本复杂度上相较于经典同类算法才能实现完整的二次加速。有趣的是,我们发现由参数化量子电路(parametrized quantum circuits)推导得到的强化学习策略在这些条件下表现良好,彰显了全量子强化学习框架的优势。

英文摘要:

Understanding the power and limitations of quantum access to data in machine learning tasks is primordial to assess the potential of quantum computing in artificial intelligence. Previous works have already shown that speed-ups in learning are possible when given quantum access to reinforcement learning environments. Yet, the applicability of quantum algorithms in this setting remains very limited, notably in environments with large state and action spaces. In this work, we design quantum algorithms to train state-of-the-art reinforcement learning policies by exploiting quantum interactions with an environment. However, these algorithms only offer full quadratic speed-ups in sample complexity over their classical analogs when the trained policies satisfy some regularity conditions. Interestingly, we find that reinforcement learning policies derived from parametrized quantum circuits are well-behaved with respect to these conditions, which showcases the benefit of a fully-quantum reinforcement learning framework.

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。

↑