Super-recursive Features of Evolutionary Processes and the Models for Computational Evolution

To build the systems that can adapt to their environments and learn from their experience is a long-standing goal and today it attracts researchers from many fields. But the power of evolutionary information processing that allows biological complex systems to adapt is still out of reach. With its extraordinary evolvable capabilities to adapt in rapidly changing and extremely hostile environment, brainless problem solvers such as bacteria successfully perform emergent computation for which they have no predetermined instructions or any a priori given referential or target values toward which they could converge. It is hard to capture adequately the adaptive efficiency of bacterial ‘machines’ within algorithmic notion of computation of Church-Turing thesis (CTT). With respect to different kinds of evolutionary information processing, overviewed in this paper, and in the direction of recently suggested study of computational evolution, we find it more adequate to exploit the diversity of algorithmic universe which has been revealed by the theory of super-recursive algorithms. According to CTT , in the current computational machines there is no place for mutations (mistakes in the genetic instruction set that survive actions of repair mechanisms and become possible source of inventions) which is one of the central concepts in the biology. Hence, emergent processes of evolution cannot even start by computation of Turing machine. The superrecursive algorithms are basis of a new paradigm for computation that changes computational procedures. Computation of Inductive Turing machines based on super-recursive algorithms presents transition from terminating computation to emergent computation. They work with the finite objects and obtain the results in a finite period of time without halting. Additionally, we may identified Super-recursive features of evolutionary processes and computational evolution. 2 different characteristics in the processes of adaptation of bacteria which we discuss here as super-recursive features of evolutionary information processing. The model of evolutionary computer discussed here is the one we need to do experiments that specifically test for essential questions related to computational evolution and that could possibly lead to exploitation of physical ideas about the nature of evolution.