Discrete-state phasor neural networks

Author:

Abstract
An associative memory network with local variables assuming one of q equidistant positions on the unit circle (q-state phasors) is introduced, and its recall behavior is solved exactly for any q when the interactions are sparse and asymmetric. Such models can describe natural or artificial networks of (neuro-)biological, chemical, or electronic limit-cycle oscillators with q-fold instead of circular symmetry, or similar optical computing devices using a phase-encoded data representation.

This publication has 16 references indexed in Scilit: