Semantic Supervised Training for General Artificial Cognitive Agents

Abstract

The article describes the authors’ approach to the construction of general-level artificial cognitive agents based on the so-called «semantic supervised learning». Within this approach in accordance with the hybrid paradigm of artificial intelligence, both machine learning methods and methods of the symbolic approach («good old-fashioned artificial intelligence») are used. A description of current problems with understanding of the general meaning and context of situations in which narrow AI agents are found is presented. The definition of semantic supervised learning is given and its relationship with other machine learning methods is described. In addition, a thought experiment is presented, which shows the essence and meaning of semantic training with a teacher, which makes it possible to «educate» a general-level AI agent. It opens the opportunity to apply not only general-level knowledge about the world around the agent, but also introduce the personal experience, which, according to the authors, will lead to a full understanding of the context and, ultimately, to the construction of general artificial cognitive agents. The article also provides a possible architecture for an artificial cognitive agent that can implement semantic supervised learning. The novelty of the work lies in the authors’ approach to combining various methods of artificial intelligence within the hybrid paradigm. The relevance of the work is based on the ever-increasing interest in general-level artificial intelligence methods in science, technology, business and world politics. The article is theoretical. The article will be of interest to specialists in the field of artificial intelligence (especially in the direction of building artificial general intelligence), philosophy of consciousness, and in general to all those who are interested in up-to-date relevant information about the approaches and methods of implementing general artificial cognitive agents.