Figure 8.

Confusion matrices for various critical boundaries in the Cell Collective computed using average node activation spread measures. Each matrix corresponds to a given threshold parameter that is evaluated to predict the dynamical regime. From left to right, these are ${\left({\sigma }^2\right)}^{\prime }k$, as in the traditional in-degree approach, followed by ${\left({\sigma }^2\right)}^{\prime }{k}_e$, and $H^{\prime }{k}_e$, our proposed measures. The predicted regime is given by the horizontal labels, and the ground truth regime, as computed from the Derrida coefficient, is given by the vertical labels. The center of each predicted critical regime is taken to be the corresponding binary classification boundary in Figure 5, and the width of the predicted critical regime is chosen to maximize the accuracy of each classifier. The ground truth critical regime is defined as the range of Derrida coefficient values centered at $\delta =1$ with a width equal to the IQR of the Derrida coefficient distribution in the Cell Collective.