Fig. 3: The Fritz distribution, theoretical versus experimental.
From: Experimental nonclassicality in a causal network without assuming freedom of choice
a Ideal Fritz distribution computed by choosing \({\rho }_{AB}=(\left|HV\right\rangle -\left|VH\right\rangle )/\sqrt{2}\) (a noiseless singlet state); ΛAC = ΛBC as classically, perfectly correlated mixed states; and the ideal measurement operators described in Eq. (5). b Experimental distribution measured in an experimental run. The error bars are calculated using Poissonian statistics and are not visible in the plot. The three indexes a, b, and c indicate the measurement results, ranging from 0 to 3, corresponding to the three nodes A, B, and C, respectively. The chart bars representing the terms of the probability distribution have different colors based on the value of the outcome c.
