(Phys.org)—A pair of physicists with ETH Zurich has developed a way to use an artificial neural network to characterize the wave function of a quantum many-body system. In their paper published in the journal Science, Giuseppe Carleo and Matthias Troyer describe how they coaxed a neural network to simulate some aspects of a quantum many-body system. Michael Hush with the University of New South Wales offers a Perspectives piece on the work done by the pair in the same journal issue and also outlines the problems other researchers have faced when attempting to solve the same problem.