A paper published in Physics Review X in May shows that neural networks may be able to describe quantum systems in new and unique ways. The paper is a collaboration between researchers from the University of Maryland’s Joint Quantum Institute and Condensed Matter Theory Center and two Chinese institutions.

Image credit: E. Edwards / JQIPhysicists typically represent quantum systems as a list of numbers illustrating the likelihood that a system will be found in different quantum states. But as systems become more complex, representation becomes much more difficult. The researchers found neural networks, which are a computational model loosely based on the biological brain and used in artificial intelligence activities such as deep learning, to be surprisingly effective in representing highly entangled quantum systems.

“Recently, machine-learning techniques have been introduced to tackle several questions about the behavior of many interacting quantum particles,” the study authors say in a summary. “Success in this context relies vitally on the underlying data structures of the quantum states that are encoded in the artificial neural networks. We explore the data structures of neural-network quantum states by studying their entanglement properties with a focus on a machine-learning model known as the restricted-Boltzmann-machine (RBM) architecture.”

“Our results reveal some crucial properties of the data structures of neural-network quantum states, which pave a novel way to bridge machine learning and many-body quantum physics,” the summary continues. “We expect that our findings will spark further exciting progress in the emerging interdisciplinary field of machine-learning quantum phases of matter.”