On the A.S. Convergence of the Kohonen Algorithm with a General Neighborhood Function
Fort, Jean-Claude ; Pages, Gilles
Ann. Appl. Probab., Tome 5 (1995) no. 4, p. 1177-1216 / Harvested from Project Euclid
Some existence and stability results for the equilibrium points of the one-dimensional Kohonen self-organizing neural network with two neighbors are extended to most nonincreasing neighborhood functions. All the functions mentioned in the neural literature are included. The assumption on the stimuli distribution is weakened, too. In the multidimensional setting, we derive from a general formula various stability and instability results.
Publié le : 1995-11-14
Classification:  Self-organization,  quantization,  stochastic algorithms,  neural networks,  60J05,  60K30
@article{1177004611,
     author = {Fort, Jean-Claude and Pages, Gilles},
     title = {On the A.S. Convergence of the Kohonen Algorithm with a General Neighborhood Function},
     journal = {Ann. Appl. Probab.},
     volume = {5},
     number = {4},
     year = {1995},
     pages = { 1177-1216},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1177004611}
}
Fort, Jean-Claude; Pages, Gilles. On the A.S. Convergence of the Kohonen Algorithm with a General Neighborhood Function. Ann. Appl. Probab., Tome 5 (1995) no. 4, pp.  1177-1216. http://gdmltest.u-ga.fr/item/1177004611/