On the generalization ability of diluted perceptrons
P Kuhlmann, K. Robert Müller · Journal of Physics A Mathematical and General · 1994
A linearly separable Boolean function is teamed by a diluted perceptron with optimal stability. A different level of dilution is allowed for teacher and student perceptron. The learning algorithms used were the optimal annealed dilution and Hebbian dilution. The generalization ability, i.e. the probability to recognize a pattern which has not been learned before, is calculated in replica symmetry.