Deep neural networks classifying transfer efficiency in complex networks

Authors:
Publisher website
Google Scholar
Abstract
Transport of particles in networks is a broad interdisciplinary scientific field. Efficient transport of particles is of practical importance for communication tasks. To improve the efficiency of the transport, the quantum phenomenon of interference is of relevance. A particle that is transferred through a network interferes with itself, allowing for transfer properties that are different from non-quantum, classical, dynamics. However, it is not known in which networks quantum properties provide an advantage. It was shown in Melnikov et al. [New J. Phys., Vol. 21, 125002, 2019] and [Adv. Quantum Technol., Vol. 3, 1900115, 2020] that convolutional neural networks are well-suited for classifying which complex networks can have a fast quantum transport. Here we discuss these results and applicability of the approach.
Funding Information
  • Ministry of Science and Technology
  • Institute of Physics