Delgado Francisco, Cardoso-Isidoro Carlos
Tecnologico de Monterrey, School of Engineering and Science, Atizapan 52926, Mexico.
Tecnologico de Monterrey, School of Engineering and Science, Monterrey 64849, Mexico.
Entropy (Basel). 2023 Feb 18;25(2):376. doi: 10.3390/e25020376.
Quantum information applications emerged decades ago, initially introducing a parallel development that mimicked the approach and development of classical computer science. However, in the current decade, novel computer-science concepts were rapidly extended to the fields of quantum processing, computation, and communication. Thus, areas such as artificial intelligence, machine learning, and neural networks have their quantum versions; furthermore, the quantum brain properties of learning, analyzing, and gaining knowledge are discussed. Quantum properties of matter conglomerates have been superficially explored in such terrain; however, the settlement of organized quantum systems able to perform processing can open a new pathway in the aforementioned domains. In fact, quantum processing involves certain requisites as the settlement of copies of input information to perform differentiated processing developed far away or in situ to diversify the information stored there. Both tasks at the end provide a database of outcomes with which to perform either information matching or final global processing with at least a subset of those outcomes. When the number of processing operations and input information copies is large, parallel processing (a natural feature in quantum computation due to the superposition) becomes the most convenient approach to accelerate the database settlement of outcomes, thus affording a time advantage. In the current study, we explored certain quantum features to realize a speed-up model for the entire task of processing based on a common information input to be processed, diversified, and finally summarized to gain knowledge, either in pattern matching or global information availability. By using superposition and non-local properties, the most valuable features of quantum systems, we realized parallel local processing to set a large database of outcomes and subsequently used post-selection to perform an ending global processing or a matching of information incoming from outside. We finally analyzed the details of the entire procedure, including its affordability and performance. The quantum circuit implementation, along with tentative applications, were also discussed. Such a model could be operated between large processing technological systems using communication procedures and also on a moderately controlled quantum matter conglomerate. Certain interesting technical aspects involving the non-local control of processing via entanglement were also analyzed in detail as an associated but notable premise.
量子信息应用在几十年前就已出现,最初引入了一种平行发展模式,模仿经典计算机科学的方法和发展路径。然而,在当前这十年中,新颖的计算机科学概念迅速扩展到量子处理、计算和通信领域。因此,诸如人工智能、机器学习和神经网络等领域都有了量子版本;此外,还讨论了学习、分析和获取知识的量子大脑特性。物质聚集体的量子特性在这一领域已得到初步探索;然而,能够进行处理的有组织量子系统的建立可以在上述领域开辟一条新途径。事实上,量子处理涉及某些必要条件,比如输入信息副本的建立,以便在远处或就地进行差异化处理,从而使存储在那里的信息多样化。最终,这两项任务都提供了一个结果数据库,可用于进行信息匹配或对至少一部分结果进行最终全局处理。当处理操作和输入信息副本数量很大时,并行处理(由于叠加,这是量子计算的一个自然特性)就成为加速结果数据库建立的最便捷方法,从而获得时间优势。在当前的研究中,我们探索了某些量子特性,以实现一个基于待处理、多样化并最终汇总以获取知识(无论是模式匹配还是全局信息可用性)的通用信息输入的整个处理任务的加速模型。通过利用量子系统最有价值的特性——叠加和非局域特性,我们实现了并行局部处理以建立一个大型结果数据库,随后使用后选择来进行最终全局处理或对外部传入信息进行匹配。我们最终分析了整个过程的细节,包括其可行性和性能。还讨论了量子电路实现以及初步应用。这样一个模型既可以在使用通信程序的大型处理技术系统之间运行,也可以在适度受控的量子物质聚集体上运行。作为一个相关但值得注意的前提,还详细分析了涉及通过纠缠进行处理的非局域控制的某些有趣技术方面。