Ontophylogenetic algorithms for the synthesis of intellectual phenotypes of software agents for use in tasks of multigenerational optimization of control neurocognitive architectures

The purpose of the study is to develop methods and algorithms for the ontophylogenetic synthesis of artificial intelligence software agents based on multi-agent neurocognitive architectures that allow combining the situationality and explanatory power of reinforcement learning and the adaptive efficiency and stability of genetic algorithms. An algorithm for synthesizing the phenotypes of control systems of intelligent agents based on the data of their genotypes has been developed. A software package for simulating the processes of ontophylogenetic synthesis of multi-agent neurocognitive architectures has been also developed. Experiments were carried out to create phenotypes of intelligent agents based on the developed genotypes of control multi-agent neurocognitive architectures.