RUS  ENG
Full version
JOURNALS // Artificial Intelligence and Decision Making // Archive

Artificial Intelligence and Decision Making, 2014 Issue 2, Pages 52–62 (Mi iipr346)

Intelligent systems and technologies

The evolutionary method of cognitive development of students autonomous Agents

V. V. Stalsky


Abstract: Proposed evolutionary method of improving the system of artificial intelligence, operating in automatic mode off – agent. Agent is designed to solve scientific and engineering problems, the agent must continually educate themselves and improve. In the method used for this law of evolution – an imitation of “sexual reproduction”: agents transmit their new upgraded software agents “next generation”. Several “generations” of agents create multi-agent system (MAS) consisting of a large number of agents – “relatives”, ie already “population” breeding agents. Population evolves due to higher “mental sposbnostey” agents from generation to generation.

Keywords: agent, an evolutionary method, multi-agent system, “sexual reproduction” population.



Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024