Evolution, emergence, semiosis: components of the model for intelligent system

Abstract

The discussion of intelligent system usually starts with issues of defining intelligence as a set of skills, but always ends with specifying the mechanisms of learning. It is important to address the issue of differences and similarities between the techniques of computational/control learning processes (very similar to the processes of semiosis) and biological learning including evolution of species where the resemblance with semiosis is less obvious. We would like to attract attention to the theory of multilevel processes of evolution which are interpreted in this paper as multiresolutional processes of evolution. Novel explanations are preposed for numerous paradoxes known in the area of computational and biological learning including evolution of species. The direct linkage is demonstrated of learning processes and the development of decision-making mechanisms for single and multiple agents.