Quantum Computing Meets Machine Learning: A New Field of Quantum Machine Learning

In today's computer science, machine learning and quantum computing are two hot research fields. In machine learning, artificial neural network (ANN) as a representative method has achieved great performance improvement in the fields of vision, speech, natural language understanding, games and other applications, which has greatly promoted the development of artificial intelligence, and also made people have greater imagination space for future general intelligence. In the field of quantum computing, breakthroughs in hardware implementation have made people see that the pace of large-scale universal quantum computers is getting closer and closer, and the possible subversive changes in information processing of future quantum computers have also made people full of expectations and interest in continuous exploration.

In recent years, a research hotspot is the combination of machine learning and quantum computing. The work on quantum machine learning has emerged in a large number in the short term. In the early stage of development, the direction, style and maturity of each work are different. This slightly messy situation may bring some confusion to beginners.

Jonathan Allcock of Tencent Quantum Laboratory and Zhang Shengyu wrote the opinion article "Quantum Machine Learning" in the first issue of Quantum Computing published by National Science Review (NSR) in 2019. To introduce.

Considering the limitation of space and avoiding duplication with previous reviews, this paper will focus on introducing the field of interaction between artificial neural networks and quantum information processing. This includes not only the expansion and application of classical artificial neural networks such as Boltzmann machine, autoencoder and generative countermeasure network (GAN) in quantum information, but also the acceleration of the calculation of classical artificial neural networks by quantum computer. In this paper, the application of quantum entanglement theory in addition is also discussed. An attempt to understand classical neural networks.

The author hopes that this article will enable the readers to have a preliminary understanding of this field, promote more scholars to enter this new and expected field, and make more solid and in-depth research. (Source: Science Network)

Paper information: https://doi.org/10.1093/nsr/nwy149