초록 ▼

A mask neutral network for processing that allows an external source of control to continuously direct state transition of the neural network toward selected states and away from other states. The network, through externally controlled masking, can focus attention on selected attributes of observed

A mask neutral network for processing that allows an external source of control to continuously direct state transition of the neural network toward selected states and away from other states. The network, through externally controlled masking, can focus attention on selected attributes of observed data, solutions or results. The masking is appliciable across three major categories of networks in that it facilitates augmented recall, directed learning and constrained optimization.

대표청구항 ▼

A method for mask controlling a neural network having a plurality neurons, having summation inputs, having outputs, and having an interconnect matrix interconnecting said summation input and said outputs variously with weighted connections having weighting values that signify the strength of interco

A method for mask controlling a neural network having a plurality neurons, having summation inputs, having outputs, and having an interconnect matrix interconnecting said summation input and said outputs variously with weighted connections having weighting values that signify the strength of interconnections among the neurons, and having threshold inputs, comprising: determining goal of said neural network; setting a control algorithm to provide variable, specified importance values for accomplishment of the goal; providing the variable, specified importance values to a mask; holding in the mask the variable, specified importance values; and inputting from the mask the variable importance values to the threshold inputs thereby affecting thresholds of said neurons, controlling said neural network and causing said network to attain the goal.