forest.benchmarking.distance_measures.watrous_bounds(choi: numpy.ndarray) → Tuple[float, float]

Return the Watrous bounds for the diamond norm of a superoperator in the Choi representation.

If this is applied to the difference of two Choi representations it yields bounds on the diamond norm distance.

The bound can be found in this StackOverflow answer, although the results can also be found scattered in the literature.

Parameters:choi – dim1 by dim2 matrix (for qubits, dim = 4**Ni, where Ni is a number of qubits)