Dual processes in neural network models I. Neural dynamics versus dynamics of learning
A C C Coolen, L G V M Lenders · Journal of Physics A Mathematical and General · 1992
The dynamics of learning in an unsupervised formulation of the Kohonen model is shown to be equivalent to the dynamics of local order parameters in an attractor neural network with short-range Hebbian interactions and long-range anti-Hebbian interactions. This duality can be used to study analytically the creation of topology conserving maps.