Fig. 5: Two-dimensional representation of the final pulse vectors at 60 ns using the t-SNE algorithm.
From: Global optimization of quantum dynamics with AlphaZero deep exploration
t-SNE is a visualization method that here represents the final pulse vectors by a set of two-dimensional vectors, in such a way that if any two pulse vectors are close to each other then the coordinates of their two-dimensional representations should be relatively close to each other as well. The color scale shows the infidelity of the pulses. a GRAPE with random seeding, b AlphaZero, c The Hybrid, i.e. AlphaZero solutions after being optimized with GRAPE. In the latter case, some example high-fidelity pulses are shown.
