Conditioned adaptive behavior from Kalman filter trained recurrent networks
L.A. Feldkamp, Danil V. Prokhorov, Timothy M. Feldkamp · 2004
We demonstrate that a fixed-weight neural network can be trained with Kalman filter methods to exhibit input-output behavior that depends on which of two conditioning tasks had been performed a substantial number of time steps in the past. This behavior can also be made to survive an intervening interference task.