Neurocognitive learning algorithm for a multi-agent system for evolutionary modeling of gene expression according to PCR analysis of plants

The work is aimed at creating a methodology for using general artificial intelligence systems to manage the process of creating new plant hybrids with a given set of economically useful traits. The basic principles for creating plant simulation models based on multi-agent modeling based on enlarged conditional cell agents, the synthesis of whose behavior is carried out by the controling neurocognitive architecture, have been developed. The basic principles for creating an automatic data collection system for evolutionary machine learning of intelligent expert systems for breeding and seed production based on robotic digital phenotyping and genetic data have been developed. An algorithm has been developed for training a decentralized system for controlling the growth and development of plant simulation models based on the identification of phenogenotypic characteristics of growth and development processes determined by the expression of plant genes.