Computing Semilinear Sparse Models for Approximately Eventually Periodic Signals

Some elements of the theory and algorithmics corresponding to computation semilinear sparse models for discrete-time signals are presented. In this study, we will focus on approximately eventually periodic signals, that is, can exhibit an aperiodic behavior initial amount time, then become afterwards. The considered in study obtained by combining representation methods, linear autoregressive GRU neural network models. Firstly, each block model is fitted independently, using some reference data signal under consideration. Secondly, mixing parameters fitted, order obtain a consisting combination previously blocks, aforementioned data. Along process, representations matrices resulting computed. prototypical computational implementations presented as well.