On linear separability of random subsets of hypercube vertices
Marco Budinich · Journal of Physics A Mathematical and General · 1991
The classical Cover results on linear separability of points in R d are a milestone in neural network theory. Nevertheless they are not valid for digital input networks because in this case the points are not, in general position, vertices of a d-dimensional hypercube. The author shows here that for large d all Cover findings can be extended to this case. It is also shown that for n