The Basic Kak Neural Network with Complex Inputs

Abstract

The Kak family of neural networks is able to learn patterns quickly, and this speed of learning can be a decisive advantage over other competing models in many applications. Amongst the implementations of these networks are those using reconfigurable networks, FPGAs and optical networks. In some applications, it is useful to use complex data, and it is with that in mind that this introduction to the basic Kak network with complex inputs is being presented. The training algorithm is prescriptive and the network weights are assigned simply upon examining the inputs. The input is mapped using quaternary encoding for purpose of efficienty. This network family is part of a larger hierarchy of learning schemes that include quantum models.