I am looking into the algorithm to calculate the quantum volume of a given quantum computer. The one thing that is unclear to me is how the achievable depth is defined.
In Andrew W. Cross, Lev S. Bishop, Sarah Sheldon, Paul D. Nation, and Jay M. Gambetta Phys. Rev. A 100, 032328 (2019) [arXiv:1811.12926], they write:
We define the achievable depth $d(m)$ to be the largest $d$ such that we are confident $h_d > 2/3$.
Where $h_d$ is the probability of observing a heavy output by implementing a randomly selected depth $d$ model circuit.
Where does the value $2/3$ come from? Why does one take $2/3$? And not, e.g. $3/4$? Is it an arbitrary choice, or is there statistical meaning to it?
