[comp.ai.neural-nets] Sparse NN's using BP info please :-)

lundonj@ul.ie (05/15/91)

Dear Neural.Netters,

		  I am interested in the concept of sparse NN's using the BP
algorithm but to my knowledge little has been reported on this topic. What I
mean by sparseness of a NN is that for instance not all inputs travel to all 
hidden neurons of the hidden layer. This reduces the dimensionality of the
problem to be computed. Any thoughts or better still any useful papers that
may help me understand the dynamics of the problem better. Responses will be
posted if large enough.