Information State: A Representation for Dynamic Processes Using Information Theory
Amanda Gondim de Oliveira, Allan De Medeiros Martins, Adrião Duarte Dória Neto · 2018
Information theory (IT) is a branch of mathematics that studies information quantification. By means of the techniques of information theoretic learning (ITL), we can estimate measures such as entropy or information potential at each moment of time of a dynamic process and analyze its evolution.In view of the dynamic behavior of information measures, this work proposes a new representation of dynamical processes called Information State Representation (ISR). In this representation, the states of the system are described as a function of the information measures of the system. In order to validate this new form of representation, some experiments were carried out with videos aiming at evaluating its quality when submitted todifferent noises.