Fig. 4: Random quantum circuits.
From: Quantum process tomography with unsupervised learning and tensor networks
a One-dimensional quantum circuit with N = 10 qubits and D = 4 layers, each one consisting of random single-qubit rotations and CX gates, the latter applied in a staggered pattern between even and odd layers. b Two-dimensional random quantum circuit, where each layer applies random single-qubit rotations and CX gates according to the colored sequence shown at the bottom of the image. In the panels (c) and (d), we show the reconstruction infidelity at each epoch, respectively, for a one- and two-dimensional quantum circuit with depth D = 4, for various data set sizes M. Subplots (e) and (f) show the lowest infidelities, obtained via cross-validation on held-out data, as a function of the data set size M for different depths.
