Motorola Corporation

Stephen G Osborn

L Rodney Goke

Michael G Gallup

Robert W Seaton, Jr

Terry G Lawell

Thomas J Tomazin

A data processing system (55) and method thereof includes one or more data processors (10). Data processor (10) is capable of performing both vector operations and scalar operations. Using a single microsequencer (22), data processor (10) is capable of executing both vector instructions and scalar instructions. Data processor (10) also has a memory circuit (14) capable of storing both vector operands and scalar operands.

Data processing system and method thereof

A method and apparatus for approximating a sigmoidal response using digital circuitry for neural network computations. The digital circuitry in processing element (16) produces the neuron output signal (110) by performing a squashing operation which determines an approximation of a sigmoid function. In one form, the present invention uses digital circuitry (16) in data processor (10) to approximate a sigmoid function of a neuron (100) using a plurality of parabolas. In an alternate embodiment, the sigmoid function of neuron (100) is approximated using a quasi-log.sub.2 function.

Growing community of inventors

Austin, TX, United States of America

Stephen G Osborn