The Brain-State-in-a-Box Neural model is shown to converge to the hypercube's extreme points under mild assumptions. Absorption at other fixed points of the activity rule is shown to be avoided by a simple randomization. This is possible because, as is shown, the extreme points are the only stable equilibria. © 1988.
Greenberg, H. J. (1988). Equilibria of the brain-state-in-a-box (BSB) neural model. Neural Networks, 1(4), 323–324. https://doi.org/10.1016/0893-6080(88)90005-6