Modularity, Self-Organization and Complexity Estimation: Three Foremost Supplies toward Intelligent Information Processing

Over the past decades, new approaches based on artificial neural networks (ANN) have been proposed to process complex information and to overcome difficult problems solution as those related to optimization, modeling, decision making, control, classification, data mining, nonlinear functions (behavior) approximation etc.. However, if learning and generalization capabilities of ANN made appear a number of expectation to emerge "intelligent" information processing, it is now admitted that intelligent behavior requires more sophisticated mechanisms than those performed by these "simple" model. Advances in "neurobiology" allowed highlighting some of key mechanisms of biological intelligence. Among them one can accentuate brain\´s "modular" structure, it\´s "self-organizing" capabilities and it\´s complexity reducing ability. The main goal of this paper is to show how these foremost supplies could be embedded in intelligent artificial mechanisms in order to design intelligent information processing systems.