초록 ▼

An artificial neural system has input operational amplifiers providing differential voltage input signals to a neuron including a voltage divider network having a plurality of substantially equal resistances selectably connectable to the components of the input signals so as to define the bits of bi

An artificial neural system has input operational amplifiers providing differential voltage input signals to a neuron including a voltage divider network having a plurality of substantially equal resistances selectably connectable to the components of the input signals so as to define the bits of binary weights for each of the input signals and to generate unweighted network output voltage signals corresponding to each bit position of the weights and representing the sums of the products of each input signal and the bit at each bit position. The unweighted bit position signals are provided to a bit position weighting device which is common to all of the weights of a neuron and which weights the unweighted signals by the binary positional values of the bit positions. The unweighted bit position signals are differential signals having one component generated by reference resistances of the network, and the sign of each weight may be selected by connection of the reference resistances to one or the other of the input signal components. A preferred embodiment has only one reference resistance for each weight. This reference resistance corresponds to a sign bit position, and the reference resistances for all of the weights are connected to provide a common reference voltage component for all of the unweighted bit position signals. Differential voltage output signals from a system utilized as one artificial neural layer may be directly connected as differential voltage input signals for the voltage divider network of a system utilized as a second layer.

대표청구항 ▼

An artificial neural network comprising: input operational amplifier means for receiving a plurality of input signals and for generating therefrom a plurality of voltage signals, said voltage signals individually representing the input signals and being maintained substantially constant for each val

An artificial neural network comprising: input operational amplifier means for receiving a plurality of input signals and for generating therefrom a plurality of voltage signals, said voltage signals individually representing the input signals and being maintained substantially constant for each value of the corresponding input signal; and a neuron providing an output signal by weighting each of said voltage signals by binary weight bits in a predetermined number of bit positions, said neuron having a plurality of synapses corresponding individually to said voltage signals, each of said synapses having a plurality of synapse bit resistor means, which have substantially equal resistances and correspond individually to said bit positions, for connection to the corresponding voltage signals so as to provide synapse bit position signals corresponding individually to the values of said input signals when a weight bit at a corresponding bit position is “one”, and for disconnection from the corresponding voltage signals when said weight bit is “zero”, bit summing means for receiving said synapse bit position signals and for generating therefrom bit position sum signals representing the sums of the synapse bit position signals corresponding to each of said bit positions, and bit position weighting means for receiving said bit position sum signals and for generating said output signal therefrom so that said output signal corresponds to the sum of the products of each of said bit position sum signals and the positional values of the bit positions corresponding to the bit position sum signals, whereby said output signal corresponds to the sum of the products of the values of each of said input signals and weights represented by the weight bits of the synapses receiving the input signals.