apfiffer@admin.ogi.edu (Andy Pfiffer) (12/18/90)
I'm considering using a connectionist method for the control of a simulated "gizmo" that has O(30) controls. These parameters are required to vary over a simulated time quanta to a slowly varying input. It is not required for the entire application to operate in "real time" because it is a simulation. Any articles, books, or papers that deal with recurrent and/or time-varying outputs from neural networks would be appreciated. Thanks. -- Andy Pfiffer (503) 645-1886 apfiffer@admin.ogi.edu Formerly w/ Cogent Research, Formerly w/ Topologix, Formerly w/ Theory Center. "To determine where to resume execution upon leaving the trap handler, examine the instruction at address fir - 4.