Communicative-Associative Transition to Smart Artificial Intelligence by Criteria with Help of Ensembles of Diversified Agents

Abstract

Cognitive virtual smart artificial intelligence can be formed by ensembles of diversified agents with strong artificial intelligence based on communicative-associative logic by recurring development of professional skills, increasing visual, sound, and subject, spatial and temporal sensitivity. Several diversifiable agents that try to get the same conclusion will give a more accurate result, so several diversifiable agents are combined into an ensemble. Then, based on the criteria of utility and preference, the final result is obtained based on the conclusions of diversifying agents. This approach increases accuracy. Bagging and boosting techniques are used to form ensembles. Bagging is a combination of independent diversifiable agents by averaging patterns (weighted average, majority vote, or normal average). Boosting is the construction of ensembles of diversifiable agents consistently. The idea here is that the next agent will consider the errors of the previous agent. Due to the fact that diversifiable agents take into account errors committed by previous agents, it takes less time to get to a real response. The combination of Bagging and Boosting decision-making methods allows the development of intelligent artificial intelligence by ensembles of diversified agents. Cognitive virtual smart artificial intelligence becomes smarter through the accumulated professional experience of high-tech skills, competencies and knowledge, having increased visual, sound, subject, spatial and temporal sensitivity. Standardization of strong artificial intelligence and the use of ensembles of intelligent compatible diversified agents will help to find boundaries in which smart artificial intelligence will benefit humanity and not harm.