“Efficient Complexity”: Evolutionary Perspectives on Natural Intelligence

A fundamental question in the conjunction of information theory, biophysics, bioinformatics and thermodynamics relates to the principles and process-es that guide and control the development of natural intelligence in natural environments where information about external stimuli may not be available at prior. A novel approach to the challenge of natural learning is proposed in the framework of constrained optimization where maximums of the information fitness of the internal states of the system with the states of external stimuli under the natural constraints of natural learning are associated with optimal learning. The progress of natural intelligence can be interpreted in this framework as a strategy of approximation of the solutions of the optimization problem via a traversal or “hopping” over the extrema network of the objective function, the information fitness under the natural constraints that were examined and described. Nontrivial conclusions on the relationships be-tween the complexity, variability and efficiency of the structure, or architecture of learning models made on the basis of the proposed formalism can ex-plain the effectiveness of neural networks as collaborative groups of small intelligent units in biological and artificial intelligence.