Software Environment for Studying Intelligent Anytime Heuristic Search Methods
Alexander P. Eremeev, Ilia A. Poliushkin, Nikolai A. Paniavin · 2020 V International Conference on Information Technologies in Engineering Education ( Inforino ) · 2020
Anytime algorithms are widely used in real-time intelligent systems because they offer a tradeoff between solution quality and solving time. This property of them is valuable when the first solution can be obtained much faster than obtaining an optimal solution with the classical algorithm. In this paper, we describe several frameworks for making anytime-algorithms from classical heuristic searches and introducing a C++ implementation of these algorithms useful for demonstrating to the students the concept of an anytime heuristic search, its use, properties and tuning. These algorithms are studied by students who are studying in the master's program in the field of "Applied mathematics and Informatics" when studying methods and systems of artificial intelligence (discipline "Methods and software tools for decision support").