Searching magnetic states using an unsupervised machine learning algorithm with the Heisenberg model

• DOI: 10.1103/physrevb.99.024423
• Publication Date: 2019-01-22T23:39:04Z
• Authors: H. Y. Kwon; N. J. Kim; C. K. Lee; C. Won
• Keywords: Heisenberg model; Hamiltonian (control theory); Magnetism

Magnetism is a canonical example of spontaneous symmetry breaking. The magnetic state below the Curie temperature spontaneously broken even though Hamiltonian invariant under symmetry. Recently, machine learning algorithms have been successfully utilized to study topics in physics. We applied unsupervised find ground states Heisenberg model. A fully connected neural network was used generate spin configuration from randomly coded features, and energy selected as cost be minimized during process. found that solved by process are consistent with theoretical solution. Also, we compared results those traditional computational methods algorithm provides an efficient method solve numerically.