Correlated patterns in non-monotonic graded-response perceptrons
arXiv:cond-mat/9907392 · doi:10.1103/PhysRevE.61.6027
Abstract
The optimal capacity of graded-response perceptrons storing biased and spatially correlated patterns with non-monotonic input-output relations is studied. It is shown that only the structure of the output patterns is important for the overall performance of the perceptrons.
4 pages, 4 figures