Generative modeling is a flavor of machine learning with applications ranging from computer vision to chemical design. It is expected to be one of the techniques most suited to take advantage of the additional resources provided by near-term quantum computers. Here, we implement a data-driven quantum circuit training algorithm on the canonical Bars-and-Stripes dataset using a quantum-classical hybrid machine. The training proceeds by running parameterized circuits on a trapped ion quantum computer and feeding the results to a classical optimizer. We apply two separate strategies, Particle Swarm and Bayesian optimization to this task. We show that the convergence of the quantum circuit to the target distribution depends critically on both the quantum hardware and classical optimization strategy. Our study represents the first successful training of a high-dimensional universal quantum circuit and highlights the promise and challenges associated with hybrid learning schemes.

}, doi = {10.1126/sciadv.aaw9918}, url = {https://advances.sciencemag.org/content/5/10/eaaw9918}, author = {Zhu, D. and Linke, N. M. and Benedetti, M. and Landsman, K. A. and Nguyen, N. H. and Alderete, C. H. and Perdomo-Ortiz, A. and Korda, N. and Garfoot, A. and Brecque, C. and Egan, L. and Perdomo, O. and Monroe, C.} } @article {ISI:000482579500007, title = {Two-qubit entangling gates within arbitrarily long chains of trapped ions}, journal = {Phys. Rev. A}, volume = {100}, number = {2}, year = {2019}, month = {AUG 26}, pages = {022332}, publisher = {AMER PHYSICAL SOC}, type = {Article}, abstract = {Ion trap quantum computers are based on modulating the Coulomb interaction between atomic ion qubits using external forces. However, the spectral crowding of collective motional modes could pose a challenge to the control of such interactions for large numbers of qubits. Here, we show that high-fidelity quantum gate operations are still possible with very large trapped ion crystals by using a small and fixed number of motional modes, simplifying the scaling of ion trap quantum computers. We present analytical work that shows that gate operations need not couple to the motion of distant ions, allowing parallel entangling gates with a crosstalk error that falls off as the inverse cube of the distance between the pairs. We also experimentally demonstrate high-fidelity entangling gates on a fully connected set of seventeen Yb-171(+) qubits using simple laser pulse shapes that primarily couple to just a few modes.}, issn = {2469-9926}, doi = {10.1103/PhysRevA.100.022332}, author = {Landsman, K. A. and Wu, Y. and Leung, P. H. and Zhu, D. and Linke, N. M. and Brown, K. R. and Duan, L. and Monroe, C.} }