ip 4.; > 0 cif = ow if < 0 We are currently also experimenting with updating the connection matrix in accordance with the equations suggested by Storkey (1997, 1998, 1999.) A key property of these equations is that both wages paid to, and rent paid by, each node are positively correlated to their ST

e Hopfield net attractors are the maps. This algorithm has been tested in sparse Hopfield nets, using both standard Hopfield net learning rules and Storkey's modified palimpsest learning rule ISV99I, which provides greater memory capacity in a continuous learning context. The use of key neurons turns ou

he Hopfield net attractors are the maps. This algorithm has been tested in sparse Hopfield nets, using both standard Hopfield net learning rules and Storkey’s modified palimpsest learning rule [SV99], which provides greater memory capacity in a continuous learning context. The use of key neurons turns out