Extremal jumps of circuit complexity of unitary evolutions generated by random Hamiltonians

We investigate circuit complexity of unitaries generated by time evolution of randomly chosen strongly interacting Hamiltonians in finite dimensional Hilbert spaces. Specifically, we focus on two ensembles of random generators – the so called Gaussian Unitary Ensemble (GUE) and the ensemble of diagonal Gaussian matrices conjugated by Haar random unitary transformations. In both scenarios we prove that the complexity of exp(-it H) exhibits a surprising behaviour – with high probability it reaches the maximal allowed value on the same time scale as needed to escape the neighborhood of the identity consisting of unitaries with trivial (zero) complexity. We furthermore observe similar behaviour for quantum states originating from time evolutions generated by above ensembles and for diagonal unitaries generated from the ensemble of diagonal Gaussian Hamiltonians. To establish these results we rely heavily on structural properties of the above ensembles (such as unitary invariance) and concentration of measure techniques. This gives us a much finer control over the time evolution of complexity compared to techniques previously employed in this context: high-degree moments and frame potentials.