A NOVEL ASSOCIATIVE MEMORY SYSTEM BASED ON NEWTON’S FORWARD INTERPOLATION

This paper proposes a novel high-order Associative Memory System based on the Newton’s Forward Interpolation, which is capable of implementing error-free approximations to multi-variable polynomial functions of arbitrary order. The advantages it offers over conventional CMAC-type AMS are: high-precision of learning, much smaller memory requirement without the data-collision problem and also the advantages of much less computational effort for training and faster convergence rates than that attainable with multi-layer BP neural networks.