21 Jan 2020 - Paper on Machine learning published

“Machine learning design of a trapped-ion quantum spin simulator” is published in Quantum Science and Technology 5 (2020) 024001, ( arXiv.1910.02496). We numerically demonstrate that a machine learning neural network can be used to efficiently solve a non-linear inversion problem in an analog trapped-ion quantum simulator. The method generates experimental control parameters (such as laser intensities on each ions) needed to program a quantum simulator, in particular, to engineer an arbitrary spin lattice geometry in one, two, and three dimensions using a linear chain of ions!