APPROACHES AND METHODS OF DECISION-MAKING IN ENVIRONMENTS WITH INCOMPLETE AND UNCERTAIN INFORMATION

Abstract

Artificial intelligence (AI) plays an increasingly important role in decision-making in the modern world, providing effective solutions in various spheres of activity. One of the most important areas of decision-making for the functioning of modern society is the navigation and safety of autonomous vehicles and cyber security.

With the growing complexity and scale of the environments in which these artificial intelligence systems work, new challenges arise: yes, in many cases, autonomous agents must interact with dynamic and multifactorial environments, where an accurate description of all the factors necessary for the agent's work is often impossible. This requires the development and application of special approaches and algorithms capable of adapting to unpredictable conditions and making decisions based on incomplete or inaccurate information.

In the future, the scope and role of using artificial intelligence for decision-making will continue to grow, which will lead to even greater complexity of the environments in which these systems operate. This indicates the great potential of scientific research in this direction.

In the modern world, tasks often arise in which it is necessary to make decisions based on incomplete or uncertain information. This article provides an overview of some modern decision-making methods and approaches, their strengths and weaknesses, application features, and possibilities for integration into other industries.

The review of methods and approaches has shown that methods based on reinforcement learning are the most versatile, effective and have the potential for further improvement. Possible solutions for decision-making with planning are also shown.

The purpose of this paper is to review approaches and methods for solving decision-making problems with incomplete or uncertain information in various fields, to determine the feasibility of using them based on the requirements of the application area, and to analyze their flexibility and versatility.