Symmetry in Classification Scheme of DHNN with Even Classification
Dong Sheng Ji · Journal of Xiamen University · 2000
When studying on the classification scheme of DHNN with even classification, we propose to convert the Hamming distance of two attractors into the distance of two corresponding dots in Euclidean space. In this way,we can construct a geometric graph called attractor graph which can describes the relationship of the attractors for the network, so that we can find the set of patterns which can be stored in a network with even classification through the attractor graph.