The foundation and application of quantum machine learning: a concise literature review

http://104.196.114.195/pdfs/orduz2021quantum.pdf

In recent years , machine learning （ML） Algorithm has always been an important research object . The idea that the algorithm can be improved to obtain better generalization accuracy and efficiency , stay ML Many sub disciplines have been opened in [66]. One discipline studies the possibility of using quantum theory to gain the advantages of learning algorithms and realize the so-called quantum advantages [53]. One of the ways to realize quantum advantage is to express data through quantum tools , For example, using Hamilton's theory [30]; By inferring from multiple data inputs and multiple models [60], At the same time, learn from a large amount of data . By using recent advances in quantum numerical optimization , Quantum advantage can also be realized , This may be used in calculations similar to gradient descent [7]. However , For machine learning scientists transitioning to quantum computing , If there is no additional training or preparation , Even understand and apply some concepts behind quantum computing , May be particularly difficult .

The purpose of this article is to collect and introduce quantum mechanics in a coherent way [39,40] Some of the most relevant basic concepts and the necessary mathematical concepts behind them , Some more ML Concept , In order to facilitate the reader to establish the necessary contact , Also identify some applications 、 Algorithms and other fields surrounding this new discipline called quantum machine learning . The structure of this review article is as follows . The first 2 Section introduces quantum mechanics 、 Quantum computing 、ML Basic concepts of , And quantum machine learning about paradigms . The first 3 section , We left some QML Applications that will develop in the next few years ; And opinions and comments . The first 4 Section is the conclusion .